The use of celebrity endorsers
is a popular executional device, but it
is not without risk. The authors report three studies examining how negative
information about a celebrity can affect the brand the celebrity endorses.
Using an associative network model of memory as a theoretical framework,
they considered four moderating variables: the size of the association
set for the brand, the size of the association set for the celebrity, the
timing of the negative celebrity information, and the strength of the associative
link between the brand and the celebrity. In the first two studies, they
used a fictitious but realistic celebrity endorser and in the third they
used an actual celebrity. Negative information about a celebrity resulted
in a decline in attitude toward the endorsed brand only for the fictitious
celebrity. That general relationship was moderated in varying degrees by
association set size, timing of the negative information, and the strength
of the link between brand and celebrity.
The use of celebrity endorsers
is prevalent in advertising. In addition
to the intuitive arguments that rationalize the practice, academic researchers
have mounted empirical evidence to demonstrate the benefits of product
endorsements. Atkin and Block (1983), for example, found that the use of
a celebrity in beer advertising led to more favorable ad ratings and more
positive product evaluations. Freiden (1984) tested four types of endorsers
(celebrity, CEO, expert, and typical consumer) and determined that in comparison
with other endorser types, the celebrity endorser scored particularly well
on dimensions such as trustworthiness, believability, persuasiveness, and
likeability. In a test of the matchup hypothesis, Kamins (1990) demonstrated
that the positive impact of a celebrity endorser depends in part on proper
fit between the celebrity and the product. Some evidence even suggests
that Wall Street values the use of celebrity endorsers-Agrawal and Kamakura's
(1995) analysis of stock price movements showed that press releases announcing
celebrity endorsement contracts resulted, on average, in a .44% excess
Our research differs from prior
research by examining the impact that negative
information about a celebrity might have on consumer evaluations of endorsed
brands. Practitioners hope their target audience's positive feelings toward
a chosen celebrity will transfer to the endorsed brand or will otherwise
enhance the brand's standing. What happens, however, if either during or
after an advertising campaign negative information about the celebrity
Widely publicized incidents (e.g.,
O.J. Simpson's indictment and later
acquittal on murder charges, and Pepsi Cola's series of debacles with three
tarnished celebrities-Mike Tyson, Madonna, and Michael Jackson) suggest
that celebrity endorsers may at times become liabilities to the brands
they endorse. The fear of potential celebrity scandals has given rise to
a minitrend toward using deceased celebrities-individuals who posthumously
can no longer engage in behaviors that might bring embarrassment and injury
to the brands with which they are linked (Goldman 1994; Lefton 1994; Miller
1993). Observation also reveals an increase in the use of animated characters
as product endorsers. Callcott and Lee's (1994) content analysis determined
that 28% of animated advertisements contained cartoon-character celebrities
such as Bugs Bunny and Mickey Mouse. Like deceased celebrities, those characters
are generally immune to negative publicity.
No research with which we are familiar
has directly tested the possible
effects of negative information about celebrities on consumers' evaluations
of endorsed brands, though two recent studies have touched on the issue.
Langmeyer and Shank (1993) demonstrated a positive relationship between
people's perceptions of a celebrity (Madonna) and perception of a nonprofit
agency (Mothers Against Drunk Driving). For subjects who had a positive
(negative) image of Madonna, perceptions of MADD became more positive (negative)
after it was paired with Madonna. Tripp, Jensen, and Carlson (1994) found,
among other results, that the effect of a celebrity endorsing multiple
products is to reduce the celebrity's credibility and likeability and to
lower attitudes toward the ad itself. Though both sets of findings are
interesting, neither study examined the specific issues addressed in our
research. We report three studies examining the conditions under which
negative press about a celebrity may affect the endorsed brand.
Marketers obviously eschew nonpositive
information about their brands and
anything or anyone (such as a celebrity endorser) that is associated negatively
with their brands. They have good cause for their concern inasmuch as literature
in various psychological traditions has theorized and/or shown empirically
that negative information has disproportionate influence on consumers'
beliefs and evaluative judgments. Mizerski (1982) has done a particularly
good job of describing the underlying reasons and demonstrating from an
attribution theory perspective how negative information operates.
Using an associative network framing,
we next explain how negative information
about a celebrity can operate to lower evaluations of the advertised brand
with which the celebrity has been associated. Three studies in which we
applied associative learning principles are then reported. In the first
two we used a fictitious, albeit realistic, celebrity endorser and in the
third we used an actual celebrity.
Associative Learning and Celebrity Endorsers
Associative learning principles
are based on a conception of memory as
a network consisting of various nodes connected by associative links (Anderson
1976; Collins and Loftus 1975; Rumelhart, Hinton, and McClelland 1986).
In our research context, celebrities and brands both represent nodes, which
initially are unconnected but become linked over time through the endorsement
Feelings toward a celebrity and/or
meanings in the celebrity are expected
to transfer to the endorsed brand through their recurring association.
The repeated exposure to two stimuli results in simultaneous activation
of memory nodes representing those stimuli, building an associative link
between the two nodes (Domjan and Burkhard 1986; Klein 1991; Martindale
1991; Rumelhart, Hinton, and McClelland 1986). After an associative link
has been forged between an advertised brand and its celebrity endorser,
subsequent negative information about the endorser may result in a lower
evaluation of the celebrity, which in turn may reflect back to the endorsed
brand through the associative link established between the two entities.
The Impact of Negative Information
Repeated advertised pairings of
a brand and celebrity establish/modify
the pattern of connectivity by strengthening the associative link between
them (Anderson 1976, 1983b; Berger and Mitchell 1989; Domjan and Burkhard
1986; Fazio, Powell, and Williams 1989; Furstenberg, Sebrechts, and Seamon
1987; Klein 1991; Martindale 1991). The concept of association sets is
especially insightful for understanding that process. An association set
represents the preexisting associates, or group of concepts, that are related
meaningfully to an object (Nelson, Schreiber, and McEvoy 1992). Both the
celebrity and the brand represent nodes connected to other nodes based
on experiences with the brand and the celebrity. Those connections represent
the association set for the celebrity and brand. For example, for some
people, the association set for Michael Jordan might consist of "Chicago
Bulls," "University of North Carolina," "basketball," "baseball," "Nike,"
"Dream Team," "Gatorade," and so forth. Association sets also include individuals'
attitudes toward objects such as brands and celebrities (cf. Berger and
Mitchell 1989; Fazio, Powell, and Williams 1989; Fazio et al. 1986; Judd
et al. 1991; Noffsinger, Pellegrini, and Burnell 1983).
When a consumer thinks about a
brand, the link with the celebrity node
is animated to a certain level through spreading activation (Anderson 1983a).
The joint activation of brand and celebrity provides a path over which
one's evaluation of the celebrity has an opportunity to transfer to the
brand. The key to the process is the simultaneous activation of the brand
and celebrity nodes. Negative information about the celebrity activates
the celebrity node, which then activates the brand node to some degree
and allows reduced evaluation of the celebrity to transfer to the brand.
Studies by Noffsinger et al. (1983) and Judd et al. (1991) provide empirical
evidence demonstrating that attitudes can be affected in such a way. The
preceding discussion suggests the following general hypothesis.
H1: Given a sufficiently strong
associative link between a celebrity and
a brand, subsequent negative information about the celebrity results in
lowered evaluations of the brand.
Moderating Role of Association Set Size
Repeated pairings of celebrity
endorser and brand should facilitate celebrity
and brand becoming part of each other's association set. (For example,
what celebrities come to mind when you see the brand names Jello and Nike?)
Thinking of the brand will increase activation of the celebrity node, and
thinking of the celebrity will increase activation of the brand node. However,
thinking of a celebrity may not be as likely to activate an associated
brand node as thinking of a brand is to activate an associated celebrity
node. The level of activation of any one particular node depends in part
on the number of competing nodes associated with that node. In the marketplace,
consumers may have a larger association set for a certain celebrity than
they have for a particular brand. In that case, thinking of the brand is
more likely to activate the celebrity node than thinking about the celebrity
is to activate the brand node.
As the size of an association set
for a given concept increases, the likelihood
of any given associated node also being activated is reduced; the greater
the number of concepts activated, the less intensively each will be activated
(Anderson 1983a, b; Collins and Loftus 1975; Nelson, Bajo, and Casanueva
1985). The learning of additional facts about a concept creates competition
to take strength away from already known facts (Anderson 1983a), and activation
of a set of nodes can inhibit the activation of related nodes (Martindale
1991). For example, consider a person about whom you know two things: he
plays the violin and drives a sports car. Because only those two concepts
are associated with this person in your mind, when thoughts of that individual
come to mind, the two associated concepts (plays violin, drives sports
car) are fairly likely to be activated spontaneously to a reasonably high
degree. If over time you learn more about the person (has two children,
works at a hospital, served in the Navy, etc.), then all else being equal,
the initial two nodes (plays violin and drives sports car) become less
likely to be as highly activated. The basic principle is known as the fan
effect (Wang, Seidman, and Reggia 1988).
Though competing nodes may reduce
activation of any one target node, the
target node may still activate strongly enough to come into working memoryprovided
the association between the originating node and the associated target
node is strong (Martindale 1991). In other words, "plays violin" may be
associated so strongly with a friend that learning additional facts, though
reducing the level of activation of "plays violin," does not reduce it
Consider now celebrities and brands.
A given brand evokes a particular
association set consisting in part of information about the brand as well
as association with the celebrity endorser. As the size of the association
set for the brand increases, the level of activation of the associated
celebrity node is likely to decrease, potentially tempering the impact
of negative celebrity information on the brand. By similar logic, as the
size of the celebrity association set increases, the level of activation
of the associated brand node is likely to decrease. Hence, we propose two
H2a: As the size of the brand association
set increases, the effect on
brand evaluations of negative celebrity information decreases.
H2b: As the size of the celebrity
association set increases, the effect
on brand evaluations of negative celebrity information decreases.
In the first experiment we manipulated
brand and celebrity association
set sizes at two levels (relatively small or large set sizes) and whether
or not subjects received negative information about the advertised brand's
celebrity endorser. Of particular interest was the change in subjects'
evaluations of the advertised brand based on the negative information about
the endorser. Subjects' brand evaluations were measured at the beginning
of the study (before introduction of negative information about the celebrity)
as well as later in the study (after introduction of negative information
about the celebrity). In sum, the study had a 2^sup 4^ mixed factorial
design consisting of three between-subjects factors (brand set size, celebrity
set size, presence/absence of negative information) and one within-subjects
factor (time of dependent variable measurement).
Choice of Brand and Celebrity
To minimize preexisting knowledge
and affect due to prior exposure and
familiarity, we used a fictitious brand from a product category relevant
to our student subject pool. On the basis of pretesting, the racing bicycles
product category and the brand name Avenix were selected. (In the context
of the experiment, Avenix was a French brand name and was pronounced phonetically
to subjects as Ah-ven-ee.) A fictional celebrity was created to overcome
potential problems with using a real celebrity. Though celebrities are
by definition well known and generally popular, liking of a celebrity can
vary considerably even in a relatively homogeneous population such as students.
High variation in subjects' evaluation of a celebrity, even though randomly
distributed across experimental conditions, could cause an error variance
that swamps treatment effects. A second problem overcome by the use of
a fictional celebrity relates to the association set size manipulation.
Ideally, one might use two real celebrities who differ only in the size
of their association sets. Practically, however, that is impossible. Any
two celebrities may indeed differ in the size of their association sets,
but, regardless of how closely matched, the two celebrities would also
vary in other uncontrollable respects.
On the basis of pretesting, French
Olympic cyclist Pierre Varnay was selected
as the celebrity endorser for Avenix racing bicycles. Varnay is fictional,
but subjects were not aware of that fact and not a single subject questioned
the authenticity of Varnay or the study ruse that incarnated him. Note
that, to overcome the potential liabilities associated with using a created
celebrity (such as reduced ecological validity), we used a real celebrity
in study 3.
Size of the Association Sets
Subjects in the small association
set condition received two pieces of
information each about the celebrity and about the brand, and subjects
in the large association set condition received six items of information
each for the celebrity and the brand. For the celebrity manipulation, the
following six items were presented in the large set size condition.
1. Pierre Varnay was the winner of the 1991 European Invitational.
2. Varnay has competed in the prestigious Tour de France three times.
3. Varnay is a vegetarian.
4. Varnay is one of the few left-handed cyclists.
5. Varnay's family has always worked on a farm.
6. Varnay was born near Normandy, France.
The following six information items
were used in the large brand association
set size condition.
1. Avenix tires self-seal if punctured.
2. Avenix has quick-release hubs.
3. Avenix comes with a one-year warranty.
4. Avenix has 10 speeds (gears).
5. Avenix is made just outside Paris.
6. Avenix was designed by a graduate
of the Sorbonne, a French university.
Only the first two pieces of listed
information for celebrity and brand
were given to subjects in the small set size conditions. Pretesting determined
that attitudes toward the brand and endorser were intentionally equivalent
across the two set sizes.
After 15 student subjects were
dropped because of incomplete responses
(6) or demand awareness (9), responses from 283 subjects remained for analysis.
Three to 14 subjects participated in each session, with seven or eight
as the modal attendance.
Background and Study Ruse. Subjects
were recruited primarily from introductory
marketing courses and offered extra credit for their participation. The
experimenter informed them that a major bicycle manufacturer was interested
in their opinions about an advertising campaign for the new Avenix. That
ruse served to rationalize subjects' (1) viewing the ad campaign that was
designed to build an association between Avenix and Pierre Varnay, (2)
reading articles (including the negative information) about Pierre Varnay,
and (3) completing both pre and post measures of attitude toward Avenix.
Association Sets. Prior to viewing
the ad campaign, subjects were told
that they may not be familiar with Pierre Varnay, a French Olympic cyclist
who endorses Avenix. (One might conjecture that such an announcement encouraged
subjects to focus unduly on the ad campaign and the subsequent negative
information. However, as is apparent from our subsequent presentation of
findings, no demand-artifact explanation could possibly account for the
pattern of results obtained.) Subjects then were given either two or six
pieces of background information about Varnay and either two or six pieces
of information about Avenix. At the end of the study, written instructions
asked subjects to recall each fact through prompting questions such as
"How long is the Avenix warranty?" and "What race did Varnay win?" The
subjects recalled 91.5 to 100% of the facts, with an average of 96.4% across
conditions, thus indicating that they had learned the necessary association
Initial Evaluation of Avenix. The
initial evaluation was completed before
subjects were exposed to the ad campaign and before they read articles
that included the negative information about endorser Varnay. An initial
lead-in question ("My overall feeling about Avenix bicycles is...") was
followed by three 9-point bipolar scales (favorable/unfavorable, positive/negative,
strongly like/strongly dislike). A second lead-in question ("I think that
the Avenix bicycle is...") was followed by four additional items (good/bad,
high quality/low quality, superior/inferior, and fast/slow). The seven
items were summed to represent an overall attitude toward Avenix (Cronbach
Forging the Association. A link
between Pierre Varnay and Avenix bicycles
was established by having subjects view a series of print ads. After being
informed that the bicycle company's ad agency wanted to get college students'
reactions to the campaign, subjects received an ad folder containing 10
black and white print ads. Each ad featured the same picture of Pierre
Varnay riding an Avenix bicycle. (An actual photograph of an unknown cyclist
representing the fictitious Pierre Varnay was selected through pretesting
that identified it as the most attractive and athletic of several tested.)
The headlines, though different in each ad, all linked Pierre Varnay with
Avenix-for example, "Pierre Varnay Wins with Avenix." Each ad had minimal
body copy that conveyed in different terms Pierre Varnay's endorsement
of Avenix. The experimenter gave subjects 10 seconds for viewing/reading
each ad. Subjects then evaluated the campaign by responding to nine 7-point
bipolar scale items. Those measures were taken merely to maintain consistency
with the study ruse and are not pertinent to any hypothesis tests.
Presenting the Negative Information.
Subjects were told that there has
been some recent publicity about Varnay and that the ad agency sponsoring
the research was interested in determining students' awareness of the publicity.
Subjects then read three short magazine articles. The first two-one ostensibly
from Billboard and the other supposedly from Peoplewere about Varnay's
music and reading preferences and were included to reduce subjects' demand
awareness. The third, allegedly from Sports Illustrated, contained the
negative information. (Control group subjects received the Billboard and
People articles but not the Sports Illustrated article.) After reading
each of the three articles, subjects were asked if they had seen the article
previously, if they subscribe to the particular magazine, if they read
the magazine at least once a month, and if the article influenced how they
felt about Varnay. The purpose of those questions was to reinforce the
ruse used to present the articles.
The Sports Illustrated article
gave subjects either of two types of negative
information about Pierre Varnay: that he had engaged in steroid use or
that he had multiple driving-under-the-influence (DUI) infractions on his
record. Because the two negative information instantiations yielded virtually
identical results, we pooled the data. Obtaining similar results with two
types of negative information simply adds to study generalizability.
Memory-Clearing Task. To help ensure
that any observed effect was due to
a restructuring of longterm memory, a memory-clearing task intervened between
the negative information manipulation and measurement of the key dependent
variables. Subjects read a newspaper article on commuting and carpooling
and then were asked a variety of closedand open-ended questions about their
attitudes toward commuting, carpooling, and the use of bicycles as an alternative
to car use.
Post Evaluation of Avenix. Subjects
were told that now that they had seen
the proposed ad campaign for Avenix, the ad agency was interested in their
evaluations of Avenix. They completed the same set of measurements (i.e.,
seven 9-point bipolar scale items) as in the initial evaluation of Avenix.
Manipulation Check and Demand Awareness
Measure. As a manipulation check
on the negative information, subjects evaluated Pierre Varnay on a variety
of 9-point bipolar scales: three global evaluation measures (good/bad,
favorable/unfavorable, and positive/negative), two trust measures (trustworthy/not
trustworthy and believable/not believable), two expertise measures (knowledgeable/not
knowledgeable and qualified/not qualified), and two endorser appropriateness
measures (appropriate/inappropriate and effective/ineffective). The three
global items were summed to represent an overall attitude toward the endorser
(Cronbach alpha=.98), as were the two trustworthiness items (r=.78), the
two expertise items (r=.74), and the two appropriateness items (r=.86).
Several checks were performed to
establish that the negative information
from the Sports Illustrated article affected subjects' evaluation of Pierre
Varnay. Immediately after reading each of the articles on Varnay, subjects
were asked, among other things, whether the article had influenced their
feelings toward Varnay. Slightly more than 88% indicated that the Sports
Illustrated article had affected their feelings. A stronger manipulation
check was the evaluation subjects made of Varnay at the end of the study.
Table 1 shows that subjects who received the negative information about
Varnay had significantly less favorable evaluations of Varnay than subjects
in the control groups, who did not read the negative Sports Illustrated
To identify potential demand awareness,
subjects responded to an open-ended
question about the purpose of the study. Responses were categorized as
demand-aware if the comments identified the purpose of the study as determining
how the negative information about the celebrity would influence people's
feelings about the brand. Of the 196 subjects in the four negative information
conditions, nine (4.6%) were classified as demand-aware and hence dropped
from subsequent analysis.
Effect of Negative Information.
H1 suggests that given a sufficiently strong
associative link between the celebrity and the brand, negative information
about the celebrity will lower brand evaluations. As subjects had evaluated
Avenix twice (before and after reading articles about Avenix), repeated
measures MANOVA was used to examine whether the change in Avenix evaluations
was greater in the negative information conditions than in the control
conditions. Table 2 reports descriptive statistics and focused contrasts
between experimental conditions and their corresponding controls.
The pre and post evaluations of
Avenix averaged across the four negative
information conditions were 46.9 and 45.9, respectively, indicating that
attitudes toward Avenix declined after the negative information about the
endorser. Comparatively, a moderately positive effect of the advertising
information is evident in the control conditions with average pre and post
evaluations of 46.5 and 47.3, respectively. The within-subjects effect
for the interaction between time of measurement (pre vs. post) and condition
(negative information vs. control) is significant (F1,276= 10.1;
p<.01), supporting the expectation in H1 that negative information about
the celebrity lowers evaluations of the advertised brand.
Moderating Role of Brand and Celebrity
Association Set Sizes. H2a and H2b
propose that the impact of the negative celebrity information on evaluations
of the endorsed brand should be greater for smaller (vs. larger) association
sets. H2a speaks to the moderating role of brand set size and predicts
a time of evaluation by brand set size (with collapsing across celebrity
set size) by negative information interaction; however, the test for that
interaction is not significant (F1,279 = 1.0, p=.31). The test of H2b
on the moderating role of celebrity set size (with collapsing across brand
set size) is also nonsignificant (F1,279=2.3, p=. 13).
Though H2a and H2b are not supported,
separate contrast tests for the four celebrity/brand set size
conditions show a statistically significant attitude change toward Avenix in the small
brand set/small celebrity set/negative information condition versus its
matching control group (F =14.96; p< .01).The finding demonstrates that
the effect of negative celebrity information on the endorsed brand is greatest
when both the brand and the celebrity set sizes are relatively small; comparatively,
no other contrasts between the respective treatment and control conditions
are statistically significant.
As supported by the logic of associative
network models, our results indicate
that activation of negative information about a celebrity can have an adverse
effect-through lowered brand evaluations-on the endorsed brand with which
that celebrity is associated. Negative information had a strong effect
when the association set sizes were small for both the brand and the celebrity.
However, when either the brand or the celebrity had larger association
sets, negative celebrity publicity did not have a significant detrimental
effect on the endorsed brand. That crucial finding suggests that negative
information about a celebrity may be problematic for the associated brand
only when consumers have scant association sets, or knowledge structures,
for both brand and celebrity. When knowledge structures for brand and/or
celebrity are more fully developed, a brand may be somewhat insulated from
negative press about the endorsing celebrity.
The third study had two objectives:
(1) to examine the generalizability
to a real celebrity of the effect of negative celebrity information on
the endorsed brand and (2) to examine whether the order in which the endorser
and the brand are evaluated moderates the effect of the negative celebrity
information. In studies 1 and 2 subjects evaluated the celebrity after
evaluating the brand. The evaluation of the brand in study 3 was manipulated
as either before or after the celebrity evaluation. Study 3 had a 2^sup
3^ mixed factorial design. The between-subjects factors were order of the
brand and celebrity evaluations and presence/absence of negative information.
The within-subjects factor, as in the first two studies, was the time of
the dependent variable measurement. Use of a Real Celebrity
The use of a fictitious celebrity
in studies 1 and 2 allowed for considerable
experimental control. Control was particularly important in study 1, which
manipulated size of the celebrity association set. The association sets
for real celebrities naturally vary from consumer to consumer, rendering
any experimental manipulation of association set problematic. Though the
subjects found the experimental ruse, the background information on Varnay,
and the overall experimental environment to be realistic, we wondered whether
negative celebrity information would affect the endorsed product in the
case of a real and wellknown endorser.
The rationale for expecting negative
celebrity information to transfer
to the endorsed brand was developed previously, but would the use of a
real celebrity yield stronger or weaker effects? An argument in favor of
stronger effects is that the negative information might have greater salience
or memorability when it is about someone people already know. An offsetting
argument is that people have larger association sets for real than for
experimentally created celebrities and, as shown in study 1, association
set size can influence the impact of negative information on the endorsed
brand. Hence, the effect on brand evaluations may be weaker for a real
celebrity than for the created celebrity used in studies 1 and 2. We hypothesized,
as in Hi involving a fictional celebrity, that negative information about
a real celebrity would lead to lowered brand evaluations. We formed no
hypothesis for whether the effect of negative information about a real
celebrity would be stronger or weaker than the effect found in studies
1 and 2.
H5: Given a sufficiently strong
associative link between a real celebrity
and a brand, subsequent negative information about the celebrity lowers
Order of Brand/Endorser Evaluation
In studies 1 and 2 subjects evaluated
the brand before evaluating the celebrity.
That ordering is ecologically valid inasmuch as consumers typically make
choices at the point of purchase without celebrity information, either
positive or negative. Any impact of celebrity information on brand evaluation
must derive from long-term memory. In study 3 we manipulated the ordering
of celebrity and brand evaluations. Our reasoning was that having subjects
evaluate the celebrity first would effectively serve to prime the negative
celebrity information and its linkage with the endorsed brand immediately
prior to subjects' evaluation of the endorsed brand. We therefore sacrificed
ecological validity of the ordering for the offsetting advantage of potentially
H6: Negative celebrity information
has greater impact on brand evaluations
when the celebrity is evaluated before the brand than it does when the
celebrity is evaluated after the brand.
The Avenix bicycle was the endorsed
product. A well-known American cyclist
(Greg LeMond) was used as the endorser for Avenix. The picture in the ads
in studies 1 and 2 portrayed a very athletic cyclist riding a racing bicycle.
Because the cyclist was wearing a helmet and his face was not clearly visible,
we used the same ads in study 3 and merely substituted the name Greg LeMond
for the name Pierre Varnay in the headline and short copy. The fictitious
Sports Illustrated article provided the negative information induction
(steroid usage). Size of the brand and celebrity association sets was held
constant at three items of information each. Three facts about Avenix (drawn
from study 1) were presented to the subjects. Three true facts provided
the association set for Greg LeMond, namely that Greg is an avid outdoorsman,
who grew up in California, and has won the Tour de France three times.
Subjects and Procedures. Ninety-seven
undergraduate students completed
the study, but two were later classified as demand-aware and dropped. Procedures
were very similar to those used in the first two studies; specifically,
subjects (1) read background information about the alleged purpose of the
study, (2) learned the association sets for Avenix and LeMond, (3) completed
an initial (pre) evaluation of Avenix, (4) viewed the Avenix ad campaign,
(5) read the fictitious LeMond magazine articles, (6) performed a memory-clearing
task, (7a) completed a post evaluation of Avenix or (7b) completed an evaluation
of Greg LeMond, (8a) completed an evaluation of Greg LeMond or (8b) completed
a post evaluation of Avenix, (9) answered diagnostic questions, and (10)
were debriefed at the end of the session and told that none of the materials
used in the study should be taken to imply that Greg LeMond, in actuality,
has ever used steroids.
We performed several checks to
establish that the negative information
from the Sports Illustrated article adversely affected subjects' evaluation
of Greg LeMond. As in the previous studies, after reading the articles
about LeMond, subjects were asked whether their feelings toward him had
changed. Approximately 81% indicated that the Sports Illustrated article
had affected their feelings about LeMond. Table 5 shows that subjects who
received the negative information about LeMond had significantly less favorable
evaluations of LeMond than subjects in the control groups who did not read
the negative Sports Illustrated article.
Tests of Hypothesized Effects
Effect of Negative Celebrity Information.
H5 examines whether the deleterious
effect of negative celebrity information on the endorsed brand, as found
in studies 1 and 2, holds in the case of a real celebrity. The average
pre and post evaluations of Avenix for the negative information condition
are 47.2 and 46.7, for a decrease of .5. For the control condition, the
pre and post evaluation averages are 46.4 and 47.1, for an increase of
.7. The pattern of results is consistent with expectations, but the interaction
between type of information (negative or not) and time of evaluation (pre
vs. post) is not significant (F1,91=1.97; p=.16).
Effect of Order of Brand/Endorser
Evaluation. H6 posits that the effect
of negative celebrity information on an endorsed brand is stronger when
subjects evaluate the celebrity before evaluating the brand. Table 6 shows
that the change in evaluation of Avenix is -.4 for subjects in the negative
information group who evaluated LeMond first, whereas the change is 1.0
for the corresponding control group (overall difference 1.4). The change
in Avenix evaluations is -.7 for subjects in the negative information group
who evaluated Avenix first, but is .5 for the control group (overall difference
1.2). The information (negative or control) by order of evaluation (celebrity
or brand first) by time of evaluation (pre or post) interaction is not
significant (F1,91 = 0.16; p=.70).
Negative celebrity information
had a somewhat weaker effect than it did
in the two studies with the fictitious Pierre Varnay. One plausible explanation
is that subjects had richer association sets for LeMond than for the fictitious
Varnay, which may have diluted the impact of the negative information.
A comparison of Table 5 with Tables 1 and 3 adds support for such an interpretation
as the negative Sports Illustrated article had a greater effect on subjects'
evaluations of Varnay than on their evaluations of LeMond.
We also examined the possible effect
of the order in which subjects evaluate
the brand and the celebrity. One might expect a greater effect of negative
celebrity information on the endorsed brand when the celebrity is evaluated
just before the brand. However, as order of brand/celebrity evaluation
is not significant, negative celebrity information appears to have been
sufficiently accessible, without artifactual priming, to influence evaluations
of the brand with which the celebrity was linked.
Our research broadens the theoretical
domains used in understanding the
processes involved in celebrity endorsements by considering celebrity issues
within the context of an associative memory framework. Indeed, the desired
outcome from the use of a celebrity endorser is a strong and positive associative
link between the brand and the celebrity as a means of enhancing the brand's
equity (Keller 1993). Though advertisers expect the associative link between
the brand and the celebrity to transfer positive feelings about the celebrity
to the endorsed brand, our research shows that a lowered evaluation of
the celebrity can lower brand evaluations.
Study 1 shows that in the small
brand/small celebrity association set size
condition-albeit in no other condition-the negative information about Varnay
attenuated subjects' attitudes toward Avenix. That finding is consistent
with the results of fan effect studies (e.g., Anderson and Reder 1987)
demonstrating how activation of a particular node increases as the number
of competing nodes decreases.
The results from study 2 replicate
those from study 1. Additionally, the
data show that negative celebrity information presented prior to the brand/celebrity
pairing had a greater effect on the endorsed brand than negative celebrity
information presented after the pairing. However, the findings fail to
support the prediction that the effect of negative celebrity information
on the endorsed brand would be greater when there is a strong (rather than
weak) associative link between the brand and the celebrity.
The results of study 3 do not support
the prior studies' findings about
the deleterious effects of negative celebrity information on the endorsed
brand. We speculate that subjects had a fuller association set for Greg
LeMond than for the fictitious Pierre Varnay, which may have blunted the
effect of the negative celebrity information.
Implications and Limits
Use of celebrity endorsers has
been an enduring practice in advertising,
but our studies illustrate its potential risk. The advertiser who chooses
to use a celebrity has no control over the celebrity's future behavior.
Any negative news about a celebrity may reduce the celebrity's allure,
and therefore the appeal of the brand the celebrity has endorsed. The risk
is potentially great for new or unfamiliar brands for which the association
set is relatively scant and for which the celebrity is essentially the
primary attribute on which consumers form evaluations of the brand. Negative
celebrity information may have a much greater effect on such brands than
it does on familiar, established brands. Also at great risk are brands
closely tied to a specific celebrity, as study 2 provides suggestive directional
evidence that the effect of negative celebrity information is greater when
the associative link between the brand and celebrity is strong.
Our research has several limitations.
One is the compressed time in which
the phenomenon was examined. Each complete experiment-which involved learning
about the brand and the celebrity, developing the association between the
brand and the celebrity, and reacting to exposure to the negative celebrity
information-was done in 45 to 50 minutes. In the marketplace, the process
would occur over weeks, if not years.
Another limitation is the laboratory
setting. Because the marketplace can
be both more intense and less intense than the laboratory, the results
from our laboratory studies may not generalize to the marketplace. Celebrities
in the marketplace often evoke a much richer set of feelings and responses
than a fictitious celebrity used in laboratory research. Strong associations
between a celebrity and a brand are established over years of advertising
exposure, and individual celebrities often endorse multiple brands. Because
of the higher salience of an actual celebrity, negative celebrity information
might have more impact for an actual than for a created celebrity. Negative
information about Michael Jordan, for example, may be more startling, interesting,
and meaningful than negative information about either the fictitious Pierre
Varnay or the less famous Greg LeMond. In many ways, the marketplace can
be more intense than the laboratory, and any effect found in the laboratory
also would be expected to occur in the marketplace. However, certain characteristics
make the marketplace seem less intense than the laboratory. In the marketplace
consumers often face communication clutter, celebrities (such as Michael
Jordan and Tiger Woods) endorsing multiple brands, brands employing multiple
celebrities (such as the well-known milk mustache campaign), and sources
of information about celebrities that vary widely in credibility. All of
those factors may mitigate the impact of negative celebrity information
and limit the generalizability of a laboratory study.
A third limitation of our research
is the use of a fictitious brand (Avenix).
Ecologically, results from our experiments could at the limit be generalized
to the subset of new or relatively unknown brands for which consumers'
knowledge structures are scant. Our use of Avenix structured a situation
in which the information about the celebrity endorser was essentially the
most important ground on which subjects could form evaluations of the brand.
Further, because negative information is notable in its tendency to gain
attention and evoke cognitive effort (Taylor 1991), our experiments created
a best-case environment for the negative celebrity information to be integrated
with initial attitudes formed toward Avenix and therefore to result in
less favorable attitudes.
Though our findings are limited
in their range of applicability, the purpose
of laboratory experimentation is not to generalize findings to business
practice, but rather to test practitioners' "theories in use" and to generalize
processes to actual practice (cf. Mook 1983). Marketing practitioners apparently
operate under the assumption that negative celebrity news holds strong
potential for sullying their brands' reputations, as celebrity endorsers
are commonly discharged when negative information about them surfaces (Miciak
and Shanklin 1994).
Our findings support the practitioner's
reasons for concern, but only when
an unknown brand is advertised. Predictions derived from associative memory
models suggest that marketers have much less cause for concern when an
established brand is advertised. In the latter situation, consumers' rich
and varied cognitive structures should insulate the brand from negative
press. Understandably, advertising executives and brand managers who are
responsible for selection of a subsequently besmirched celebrity endorser
sense great urgency to dump the celebrity quickly to save face or for fear
of consumer retribution. Their actions might be fully justified for the
aforementioned reasons, but theoretical predictions based on associative
network models of memory do not necessarily indicate that negative information
about a celebrity causes serious harm to an established brand. Indeed,
Hertz apparently has not suffered any significant reduction in market share
since news emerged about O. J. Simpson's alleged connection with the murders
of his former wife and her friend, nor did Pepsi lose share after Michael
Jackson's child-molestation charge or Mike Tyson's rape conviction. Executives
undoubtedly were embarrassed in all those instances, but their well-established
brands remained relatively unscathed.
Brian D. Till is Assistant Professor of Marketing, Saint Louis University.
Terence A. Shimp is Professor of Marketing, University of South Carolina.