The last five are the articles that have to be summarized and the first one is the full instructions on the paper with exact details on how it is to be done.

I Don’t Believe It! Belief Perseverance in Attitudes Toward Celebrities

Ngoc H. Bui University of La Verne

The persistence of beliefs and feelings regarding a favorite celebrity were investigated. Source credibility was not expected to influence belief perseverance and emotional belief (feelings) perseverance toward favorite celebrities. Also, when asked to consider reasons why others would not like the celebrity, perseverance of beliefs and feelings were expected to decrease. Participants (N ! 201) from a Southern California private university completed an online survey. Contrary to the hypotheses, participants were more likely to believe the information when the media caught the celebrity and when the celebrity displayed the behavior. Also, using the consider-the-opposite strategy did not change beliefs or feelings. However, as predicted, perseverance of feelings about celebrities did vary by the source of information. The practical implications are that, although difficult, changing beliefs may be possible depending on the source of information. Limitations in the study are also discussed.

Keywords: belief perseverance, celebrities, attitudes

Interest in celebrities has been well docu- mented in research (Giles, 2000; Maltby, Day, McCutcheon, Houran, & Ashe, 2006; McCutch- eon, Lange, & Houran, 2002). According to Giles and Maltby (2004), a normal part of ad- olescence is developing admiration for and fas- cination with celebrities. This admiration and interest can impact the way we view our favor- ite celebrities in that existing positive beliefs about the celebrities may persist into young adulthood and influence behavior. For example, Boon and Lomore (2001) found that young adults are inspired by their idols (favorite ce- lebrities) to pursue creative activities and change their lifestyle in some way akin to the celebrities’ interests and lifestyle. However, headlines in the media today are filled with reports of a celebrity acting inappropriately or saying something inappropriate. Are fans’ af- fections for their favorite celebrity influenced

by these bad behaviors? Or, are fans subject to the belief perseverance phenomenon?

Emotional Beliefs Framework: Why We Can’t Let It Go

Holding on to one’s beliefs, despite contra- dictory evidence, is known as belief persever- ance (Jelalian & Miller, 1984). Several theories have been used to explain belief perseverance. For example, according to Anderson, Lepper, and Ross (1980), belief perseverance is a func- tion of the causal or explanatory scripts (i.e., cognitive schemas) that people generate to make sense of a social situation. However, these scripts are difficult to change once they are formed because they persist independent of the data that they were derived from (Anderson et al., 1980; Smith, 1982), and they continue to be accessed but never revised (Ross et al., 1975).

Although the cognitive schema explanation has been used to understand the persistence of beliefs about the self, the theoretical framework that appears to be most useful for understanding belief perseverance in attitudes toward celebri- ties is the Emotional Beliefs Framework (Mer- cer, 2010). Because beliefs related to person perception (e.g., perceiving a favorite celebrity) may have more to do with how a perceiver feels about the target person, the Emotional Beliefs

This article was published Online First December 23, 2013.

Ngoc H. Bui, Department of Psychology, University of La Verne.

Correspondence concerning this article should be ad- dressed to Ngoc H. Bui, Department of Psychology, Uni- versity of La Verne, 950 Third Street, La Verne, CA 91750. E-mail: [email protected]

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

Psychology of Popular Media Culture © 2013 American Psychological Association 2014, Vol. 3, No. 1, 38 – 48 2160-4134/14/$12.00 DOI: 10.1037/a0034916

38

Framework can help explain why celebrity at- titudes may persevere.

According to Mercer (2010), an emotional belief is a belief that is made up of and strength- ened by emotions. Emotional beliefs are based on an individual’s internal inferences, not on objective comparisons. Thus, having an emo- tional belief can mean that the belief could be wrong (Mercer, 2010). For example, profes- sional cyclist, Lance Armstrong, began the LiveSTRONG Foundation after he successfully battled testicular cancer (Lance’s Story, n.d.). His experiences inspired many people to con- tinue battling cancer and to help others suffering from the disease through donations and advo- cacy. However, in January 2013, Armstrong admitted to illegal doping while competing (and winning) in the Tour de France cycling compe- titions and the Olympics (Albergotti & O’Connell, 2013). Despite his admission, and the testimonies from teammates, his friends, and his staff regarding his illegal doping, many devoted fans of Armstrong did not waiver in their support for him, and some still view Arm- strong as an inspiration for those fighting cancer (Lopresti, 2013). These fans choose to see his positive contributions to cancer research, rather than his cheating behavior or his ban from the sport of cycling. This demonstrates that at least some fans continue to have positive emotional beliefs for Armstrong, despite his tainted repu- tation.

An emotional belief also implies that an individual has a particular point of view, and this point of view is biased by his or her emotions for that target (Mercer, 2010). How a person feels about something serves as a useful measure of his or her values or beliefs (Mercer, 2010). Therefore, if someone feels strongly about a favorite celebrity, then his or her emotions will influence what he or she believes about the person. Although no cur- rent studies have examined perceptions of celebrities using the emotional belief perse- verance model, the present study explores whether attitudes toward celebrities persist in beliefs and feelings in the face of negative or discrediting information.

Belief Perseverance and Person Perception

Belief perseverance has been shown in stud- ies on self-perception of math abilities (Lepper,

Ross, & Lau, 1986) and intelligence (Guenther & Alicke, 2008); however, there is limited re- search on belief perseverance in regard to per- son perception. Person perception involves in- terpreting and understanding others in terms of their intentions and motives, which are often unknown (Fiske, 2010). Of the few published studies that have investigated belief persever- ance and person perception, the research is over 30 years old and no studies have examined the phenomena in regard to attitudes toward celeb- rities. For example, the classic study on belief perseverance study with regard to perceptions of others asked participants to generate reasons to explain the relationship between the occupa- tional success of firefighters and the firefighters’ scores on a preference for risk measure (Ander- son, Lepper, & Ross, 1980). Even after the researchers discredited the evidence that relates occupational success and risk preference, par- ticipants continued to hold on to their explana- tions that the two variables were linked.

Belief Perseverance and Attitudes Toward Celebrities

Currently, no research to date has given attention to how our beliefs about celebrities persevere in the face of undesirable or nega- tive information from the media or other sources. Based on the Emotional Belief Framework proposed by Mercer (2010), it could be predicted that individuals would demonstrate belief perseverance about celeb- rities, even when damaging or discrediting information is presented. Mercer states that for things we care about (in this case, celeb- rity figures) we may interpret ambiguous be- haviors in ways that we prefer them to be, based on our emotions. Perhaps when our favorite celebrities are “caught” doing some- thing harmful or bad (e.g., abusing drugs, assaulting someone, or stealing) we may in- terpret the situation in a way that is based on the positive emotions we have for our favorite celebrity, which would be to try to persist in believing the celebrity did no wrong, or was “set up” by the media. Thus, according to Mercer’s (2010) Emotional Beliefs Frame- work, individuals may see or hear of their favorite celebrity’s wrongdoings but deny this information because of the strong feelings

39BELIEF PERSEVERANCE AND CELEBRITIES

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

they have previously developed for their ce- lebrity.

Measuring Perseverance

Many studies on belief perseverance use a debriefing procedure to inform the participants that the feedback or information they received before reporting their beliefs was false (Ander- son et al., 1980; Lepper et al., 1986; Ross et al., 1975). However, Wegner, Coulton, and Wen- zlaff (1985) examined belief perseverance com- paring both the traditional debriefing procedure and a briefing procedure, in which researchers informed participants before receiving feedback that the information they would obtain was in- correct. Wegner et al.’s findings show that both briefing and debriefing conditions produced be- lief perseverance effects. Wegner et al. posited that beliefs not only persevere (as seen in the debriefing condition), but also pervade later be- liefs, evidenced by the briefing condition. In the present study, a procedure similar to briefing was used. Before asking participants about whether they believed negative information about a celebrity, they were asked to consider the opposite about their celebrity. That is, they were asked to think of reasons why others would not like their favorite celebrity.

Factors Influencing Belief Perseverance

Considering the Opposite

Although belief perseverance has been demonstrated consistently in other studies (Lepper et al., 1986; Ross et al., 1975), hav- ing individuals consider opposing opinions or possibilities can reduce belief perseverance and improve judgments by reducing bias. For example, Ross et al. (1975) randomly as- signed participants to determine whether sui- cide notes were real or fake. The researchers gave participants one of three types of feed- back regarding their abilities: successful, av- erage, or failure. Ross et al. then randomly assigned participants to either a traditional debriefing procedure or a “process debriefing. ” In the traditional debriefing condition, sub- jects were told after the task that their initial feedback about their success or failure was false. Then participants were asked to esti- mate past and future performances in deter-

mining the validity of notes. Estimates corre- sponded with participants’ initial feedback, despite the discrediting of this feedback by the researchers. Specifically, those who were first told they were “successful” at identifying the suicide notes, continued to think so, and those who were first told that they were “fail- ures” at identification continued to think so. However, in the “process debriefing” condi- tion, the researchers told the participants that they received false feedback and also ex- plained to participants how belief persever- ance can influence erroneous impressions. Results showed that participants in the tradi- tional debriefing condition continued to per- sist in their beliefs more than those in the process debriefing condition.

Thoughtful processing of the information one receives can help to also reduce biased judgments. Lord, Lepper, and Preston (1984) posited that when students were asked to con- sider how an alternative outcome might be possible they would correct their view and exhibit less biased perspectives. Proponents and opponents of capital punishment were asked to consider how convincing an argu- ment was in supporting or not supporting capital punishment. Lord et al. (1984) found that when students were asked to be objective, they did not weigh the arguments objectively. However, when students were asked to con- sider the opposite, specifically that the argu- ment they heard had merit and could be pos- sible, participants believed the arguments to be more convincing, which demonstrated less judgment bias.

Findings from Ross et al. (1975) and Lord et al. (1984) show that considering the oppo- site helps to reduce belief perseverance and judgment bias. Though the strategy has not been tested on belief perseverance and person perception, in the present study it was ex- pected that considering the opposite might also influence attitudes toward celebrities. Specifically, instead of briefing individuals regarding how their beliefs are wrong (as it would be nearly impossible to dissuade each individual about the merits of liking their particular celebrity), participants were asked to think about whether there could be reasons not to like their favorite celebrity. In having participants consider the opposite, or think of reasons why others would not like the favorite

40 BUI

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

celebrity, it was expected that this approach would lessen belief perseverance for attitudes toward celebrities.

Source Credibility

Although belief perseverance can be strong, people might discount the information they re- ceive based on the credibility of the source. Source credibility is the belief that a source is authentic (Eagly, Wood, & Chaiken, 1978). Previous studies have demonstrated that source credibility can sway how individuals view in- formation (Eagly et al., 1978; Petty & Ca- cioppo, 1986). Specifically, results from a meta- analysis on source credibility, covering over 50 years and 18 different studies, demonstrated that in the presence of prior attitudes individuals did not use source credibility to be persuaded (Kumkale, Albarracin, & Seignourel, 2010). However, when individuals have not formed attitudes or do not have access to prior attitudes, source credibility becomes more crucial in per- suasion (Kumkale et al., 2010). Relevant to the present study, beliefs and attitudes about one’s favorite celebrity are previously formed from experience, and, therefore, it was expected that negative information about the celebrity should be discounted, regardless of source.

Present Study and Hypotheses

Research on belief perseveration has demon- strated the phenomenon to be strong and con- sistent (Guenther & Alicke’s, 2008; Ross et al., 1975; Wegner et al., 1985). Also, as the fasci- nation for celebrities continues to grow, there is need for research in this area to learn how celebrities influence individuals’ behaviors and thought processes. Some studies have shown that a celebrity’s negative publicity can influ- ence consumer attitudes; however, these studies include ambiguous or fictional celebrities (Thwaites, Lowe, Monkhouse, & Barnes, 2012; Till & Shimp, 1998) or famous figures that were currently in the news (Fong & Wyer, 2012). They do not examine fans’ attitudes toward their own favorite celebrities. Therefore, the present study aimed to answer several research questions: Is there belief perseverance for atti- tudes toward favorite celebrities? Do people continue to feel the same about their favorite celebrity, despite negative information pre-

sented about the celebrity (i.e., is there emo- tional belief perseverance for attitudes toward celebrities)? Does considering the opposite about the celebrity influence belief persever- ance and emotional belief perseverance? Fi- nally, does the source of the information relate to how individuals think and feel about their favorite celebrity?

Given the evidence regarding the belief per- severance effect (Ross et al., 1975) and the weak role of source credibility in persuading previously formed attitudes (Kumkale et al., 2010), it was expected that belief perseverance would be demonstrated. However, considering the opposite was expected to reduce persever- ance (Lord et al., 1984). Specifically, the fol- lowing hypotheses were tested:

1. Belief perseverance will be demonstrated in that most participants will not believe negative information presented about their favorite celebrity, regardless of the source of information.

2. Emotional belief perseverance will be demonstrated in that most participants’ feelings for the celebrity will not change when presented with negative information about their favorite celebrity, regardless of the source of information.

3. Considering the opposite about the favor- ite celebrity will relate to less belief per- severance and less emotional belief perse- verance regarding the celebrity, regardless of source of information.

Method

Participants

Of the 233 individuals recruited from a pri- vate Southern California university through research subject pools and campus student or- ganizations, 202 participants completed all measures. However, one participant was re- moved owing to not citing a favorite celebrity. The final sample consisted of 201 individuals. The 149 females and 52 males ranged in age from 18 to 58 years old, and had an average age of 21.05 (Median ! 19.5; SD ! 4.97). Partici- pants varied in ethnicity, with most of the sam- ple reporting their ethnicity as Latino/Hispanic (43.3%; n ! 87). White participants were the next largest group (30.3%; n ! 61). Most of the

41BELIEF PERSEVERANCE AND CELEBRITIES

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

participants were students (97%). The majority of participants did not work (n ! 103). Actors (39.3%; n ! 79) and musicians/singers (32.3%; n ! 65) were chosen most often as “favorite” celebrity. The 201 participants in the study named 142 different celebrities, with Taylor Swift (n ! 9), Johnny Depp (n ! 6), and Sandra Bullock (n ! 6) being the three most often named favorite celebrities, with all other celeb- rities cited less frequently (n " 5).

Measures

Participants were instructed to choose a fa- mous living person (or a famous person who died during the participant’s lifetime) whom they admired. Participants also identified what the celebrity was famous for (e.g., acting, sing- ing, writing, dancing, etc.). Participants also completed four demographic questions (i.e., gender, ethnicity, age, and work status), one item asking participants to consider reasons why other people might not like the favorite celebrity, and a 5-item measure of belief perse- verance regarding their attitudes toward their favorite celebrity (designed by the researcher).

Considering the opposite. The researcher created an item to make participants consider the opposite about their celebrity by asking re- spondents to “think of any reasons why people do not like [the participant’s] favorite celeb- rity.” Participants chose among five response options: (1) Yes, I can think of at least five or more reasons (n ! 18); (2) Yes, I can think of three or four reasons (n ! 21); (3) Yes, I can think of a couple of reasons (n ! 72); (4) Yes, I can think of at least one reason (n ! 55); and (5) No, I can’t think of any reason why anyone would not like my favorite celebrity (n ! 35). A split-median procedure divided responses into two categories, with 90 participants grouped into the first category (thinking of one or no reasons why others would dislike the favorite celebrity) and 111 participants grouped into the second category (thinking of two or more rea- sons to dislike the favorite celebrity).

Belief perseverance and source credibility measures. The five items designed by the re- searcher to measure belief perseverance and source credibility asked participants to rate their degree of belief regarding information about their favorite celebrity and whether the infor- mation changed how they felt about the celeb-

rity. Participants were instructed to use a 4-point Likert scale for the following scenarios that varied by source of information: (1) When the media reports something bad about my fa- vorite celebrity; (2) When my favorite celebrity is caught by the media acting inappropriately, rude, or crazy; (3) When my family tells me something negative about my favorite celebrity; (4) When my favorite celebrity says something inappropriate, rude, or crazy on TV, on the radio, or in a magazine; and, (5) When my friends tell me something negative about my favorite celebrity. Responses were based on a 4-point Likert scale ranging from 1 to 4, with “1” being “I don’t believe it,” “2” being “I am not sure if it is true or not, but it doesn’t change how I feel about my favorite celebrity,” “3” being “I tend to believe it but it doesn’t change how I feel about my favorite celebrity,” and “4” being “I tend to believe it and it changes how I feel about my favorite celebrity.” Although the responses were on a Likert scale, the belief categories were examined as nominal data in the analyses.

To determine belief perseverance, partici- pants’ responses were categorized into three groups based on their responses to the five be- lief perseverance items: “believe” (responses 3 and 4), “unsure” (response 2), and “don’t be- lieve” (response 1). To determine emotional belief perseverance participants’ responses were categorized into two groups: “unchanged feelings” (responses 1, 2, and 3) and “changed feelings” (response 4).

Procedure

Convenience and snowball sampling proce- dures were used to recruit participants. These sampling strategies were deemed most appro- priate because random sampling was not available to the researcher (i.e., sampling frames for all potential participants were not available), and participants needed to meet the criteria of being !18 years and have a favorite celebrity. Therefore, individuals who knew other individuals who had a favorite celebrity could recommend the study to these persons. Most of the sample was from intro- ductory and other psychology courses and were obtained between fall of 2011 and fall of 2012 through the psychology department’s research subject pool at the researcher’s uni-

42 BUI

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

versity, which is a private university in South- ern California. E-mails were also sent to uni- versity club organizations to request that the study be distributed to their members and to anyone they knew who would be interested. Participants were directed to a secure Web site link where they could access the consent form and measures. No names or student identification numbers were collected. Partic- ipants first reported their favorite celebrity and what the celebrity was famous for, and then answered the item that had them consider the number of reasons why others would not like their favorite celebrity. Following this item, they completed the 5-item measure of belief perseverance and source credibility. Completion of the measures took approxi- mately 20 min. Students could earn course credit or extra credit (depending on the in- structor) and all participants were offered the opportunity to enter a drawing to win one of four US $15 movie theater gift certificates via e-mail. A separate e-mail address was created for participants to enter the study raffle and interested participants had to just e-mail their name to the separate address to be entered. Winners were chosen at the conclusion of the study.

Results

To test the first hypothesis that most partici- pants will not believe negative information pre- sented about their favorite celebrity, regardless of the source of information, participants’ re- sponses were grouped into “believe,” “unsure,” and “don’t believe” categories for all five sce- narios varying by the source of information. Pearson goodness-of-fit chi-square analyses re-

vealed significant patterns of difference in be- liefs (p " .001). When the source of informa- tion was family, friends, or the media, the majority of participants (#50%) reported that they were “unsure” about the negative informa- tion regarding their celebrity, compared with participants who reported that they believed or did not believe the information based on these sources. However, when the scenario was that the media caught the celebrity doing something rude or inappropriate more participants reported that they believed the negative information (49. 75%) than did not believe (9.45%) or were unsure (40.80%). Also, more participants re- ported that they believed the information if the scenario depicted that the celebrity had said something inappropriate on TV, on the radio, or in a magazine (63.68%), compared with those who reported they did not believe (5.97%) or were unsure (30.35%). Contrary to the hypoth- eses, the results show that beliefs are related to the particular source of the information. See Table 1 for frequencies and for chi-square data.

As previously stated, participants’ responses were grouped into either “unchanged feelings” or “changed feelings” for all five scenarios varying by the source of information. As pre- dicted, based on Mercer’s Emotional Beliefs Framework (2010), participants showed emo- tional belief perseverance regarding their favor- ite celebrity, regardless of credibility source. Pearson goodness-of-fit chi-square analyses re- vealed significant patterns of difference in how participants felt about their celebrities, with most participants reporting “unchanged feel- ings,” despite receiving negative information about their favorite celebrity (p " .001). More than 99% of the participants (n ! 199) reported “unchanged feelings” for their favorite celebrity

Table 1 Frequencies and Chi-Square Results for Belief Perseverance (N ! 201)

Source

Don’t believe Unsure Believe Chi-square

O % O % O % $2 df

Media reports 17 8.46 140 69.65 44 21.89 124.75! 2 Family reports 47 23.38 106 52.74 48 23.88 34.06! 2 Friends’ reports 42 20.89 112 55.72 47 23.38 45.52! 2 Caught by media 19 9.45 82 40.80 100 49.75 54.00! 2 Celebrity displays behavior 12 5.97 61 30.35 128 63.68 101.22! 2

Note. O ! observed frequencies; % ! percent of total (N ! 201). ! p " .001.

43BELIEF PERSEVERANCE AND CELEBRITIES

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

when the scenarios depicted that family or friends reported negative information about the celebrity. More than 97% (n ! 196) of partic- ipants reported they had “unchanged feelings” when the scenario described that the media re- porting the celebrity had done something nega- tive. Also, when the scenario depicted that the media caught the celebrity doing something in- appropriate, #88% reported they had “un- changed feelings” for their celebrity (n ! 178). Finally, when the scenario depicted the celeb- rity saying something inappropriate, rude, or crazy on TV, on the radio, or in a maga- zine, #78% reported they had “unchanged feel- ings” for their celebrity (n ! 158). See Table 2 for frequencies and for chi-square data.

Participants were asked to consider the num- ber of reasons, if any, why others would not like the favored celebrity. As stated above, to mea- sure belief perseverance participants’ responses were grouped into “believe,” “unsure,” and “don’t believe” categories for all five scenarios varying by the source of information. Those who reported that they could think of two or more reasons to dislike the favorite celebrity (n ! 111) were compared with individuals who could think of one or no reasons to dislike the celebrity (n ! 90). Several Pearson chi-square of independence analyses were conducted to compare the two groups on belief perseverance by source. Results showed no support for the hypothesis that considering the opposite can minimize belief perseverance. In fact, the groups did not differ in their beliefs when the media reported negative information, when the media “caught” the celebrity doing some- thing inappropriate, and when the celebrity said

something inappropriately (p # .05). However, when the source of negative information was family, significantly more individuals who gen- erated two or more reasons did not believe the information (32.2%) than individuals who con- sidered only one or no reasons (16.2%), $2(2, N ! 201) ! 7.23, p ! .03. This significant pattern of difference was also found when friends were the source of information. Those who considered two or more reasons why others would not like the celebrity were more likely to not believe the information from friends (31. 1%) compared with those who considered one or no reasons (12.6%), $2(2, N ! 201) ! 10.95, p ! .004.

To measure emotional belief perseverance, responses were grouped into either “unchanged feelings” or “changed feelings” for all five sce- narios varying by the source of information. Results showed that no matter the source of the information, and regardless of the number of reasons participants generated for others not liking the favorite celebrity (i.e., considering the opposite), respondents did not change their pos- itive feelings for their favorite celebrity (p # .05). See Table 3 for frequencies data.

Discussion

The results from this exploratory study showed mixed support for the hypotheses. The findings demonstrated that beliefs about a fa- vorite celebrity changed depending on the source of information, but feelings for the ce- lebrity did not change, supporting the idea that beliefs and feelings are two separate compo- nents of belief perseverance (Mercer, 2010).

Table 2 Frequencies and Chi-Square Results for Emotional Belief Perseverance (N ! 201)

Source

Unchanged feelings Changed feelings

Observed % Observed % $2 values df

Media reports 196 97.51 5 2.47 181.50! 1 Family reports 199 99.00 2 0.99 193.08! 1 Friends’ reports 199 99.00 2 0.99 193.08! 1 Caught by media 178 88.56 23 11.39 119.53! 1 Celebrity displays behavior 158 78.61 43 21.29 65.80! 1

Note. % ! percent of total (N ! 201). ! p " .001.

44 BUI

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

Even though most participants reported that they were unsure about the negative informa- tion, these participants reported that they con- tinued to feel the same way about their favorite celebrity regardless of the source of informa- tion. Furthermore, the results showed that con- sidering the opposite does not reduce belief perseverance. In fact, when family and friends report negative information about the celebrity, individuals who can think of reasons why others don’t like the celebrity are more likely to not believe the information. Perhaps their unwill- ingness to believe the information is owing to the fact that thinking about the potential reasons others would not like the favorite celebrity might incite them to think their family and friends are “haters” of the favorite celebrity, and thus these sources would report inaccurate or false information. The findings from the present study also show that thinking about reasons why someone doesn’t like your favorite celebrity does not reduce emotional belief perseverance for that celebrity; that is, emotional belief per-

severance can withstand the consider-the- opposite strategy.

Strengths and Limitations of the Present Study

One strength of the present study is that the social psychological phenomenon of belief per- severance in the context of celebrity attitudes was explored, as this has not been examined before in the social psychological literature. An- other strength of the study is that the sample was ethnically diverse, with the majority (43. 1%) of the participants being of Latino/Hispanic descent, which allows the findings to be generalized to groups other than White/Caucasian individuals. Also, participants chose their own favorite ce- lebrity, which strengthens the findings regard- ing celebrity attitudes given that the target of the beliefs and feelings (i.e., the celebrity) was not theoretical but real.

There were several limitations to the study, including the fact that there was no clarification regarding the negative information about the celebrity. Perhaps some participants thought the worse (e.g., accused of murder, molestation, rape) or maybe some thought their celebrity was caught doing something that was a minor of- fense (e.g., leaving a restaurant without paying). The nature of the negative information was not assessed or manipulated and, therefore, could have impacted the degree in which emotional belief perseverance was demonstrated. Another limitation was that the sources of information were not specific. For example, the terms “fam- ily” and “friends” could mean different things to different participants. “Family” could mean close and immediate family members or distant relatives. “Friends” could mean best friends or mere acquaintances. Furthermore, the present study did not investigate the degree of informa- tion the participants had for their favorite celeb- rity. The amount of information that a fan may have regarding their favorite celebrity varies, and, thus, could have impacted the degree to which they would believe or not believe a source. For example, if a fan follows a celebrity Twitter feed, reads about the celebrity in mag- azines, or is a member of the celebrity’s fan club, the fan may know whether the information about the celebrity was true or false, compared with a causal fan who does not read tweets or magazine articles about the celebrity. Also, the

Table 3 Frequencies for Belief Perseverance and Consider- the-Opposite Strategy (N ! 201)

Source

Consider two or more

reasons

Consider one or no

reasons

O % O %

Media reports Not believe 7 7.8% 10 9.0% Unsure 69 76.7% 71 64.0% Believe 14 15.6% 30 27.0%

Family reportsa

Not believea 29a 32.2%a 18a 16.2%a

Unsure 41 45.6% 65 58.6% Believe 20 22.2% 28 25.2%

Friends’ reportsa

Not believea 28a 31.1%a 14a 12.6%a

Unsure 46 51.1% 66 59.5% Believe 16 17.8% 31 27.9%

Caught by media Not believe 12 13.3% 7 6.3% Unsure 39 43.3% 43 38.7% Believe 39 43.3% 61 55.0%

Celebrity displays behavior Not believe 7 7.8% 5 4.5% Unsure 31 34.4% 30 27.0% Believe 52 57.8% 76 68.5%

Note. O ! observed frequencies. a These data are significant between groups (p " .05).

45BELIEF PERSEVERANCE AND CELEBRITIES

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

belief perseverance measure created by the re- searcher could have been improved such that beliefs and feelings are measured separately, as well as specifically asking whether participants believed the information, as opposed to whether they “tend[ed] to believe” the information.

Practical Implications

In the present study, emotional belief perse- verance was demonstrated for attitudes toward favorite celebrities, regardless of source. Fur- thermore, researchers have argued that fans can become emotionally involved with their favor- ite celebrities (Maltby, Day, McCutcheon, Houran, & Ashe, 2006), and De Backer (2012) posits that celebrity images can be so powerful that people begin to think about their favorite celebrities as part of their own social network. Branch, Wilson, and Agnew (2013) also found that individuals’ satisfaction and investments to maintain a close relationship with media figures predict their commitment (i.e., persistence) to the figure. Consequently, negative information about celebrities may be discounted because fans are committed to and create emotional bonds with their favorite star.

A practical implication of the perseverance of emotional beliefs about celebrities is how com- panies and advertisers can use celebrity infor- mation to promote their stars. If fans continue to feel the same about their celebrity, despite bad publicity, can advertisers use “marred” celebri- ties in advertising or as spokespersons? There is precedent for celebrities who have kept their fans despite having had bad publicity. For ex- ample, after singer Mariah Carey received bad reviews for her movie Glitter and then sought help for “psychological problems,” she still was able to sell millions of albums, breaking records and earning accolades for her “comeback” (Werde, 2005).

Another implication of this study’s findings is that source credibility is related to belief perseverance. Specifically, celebrities, or those who deal with creating an image for celebrities, need to be aware that in the age of Twitter and Facebook, tweets and posts by celebrities must be carefully screened. Negative information portrayed through these more direct means can be as persuasive, if not more, than media re- ports. Thus, celebrities’ own accounts of their bad behavior may be powerful in changing fans’

previously positive beliefs. For example, the misdeeds of actress Lindsey Lohan reported in the media, and more importantly by her own tweets on Twitter, have been the target of ridi- cule on TV shows, magazines, and even in songs (Tauber et al., 2010). Her stints in drug rehab and jail have made it difficult for the public to look past her transgressions and focus on her acting talents, and some believe she is a bad social role model for youth (Smith, Twum, & Gielen, 2009). Therefore, understanding be- lief perseverance can be advantageous to any- one in the celebrity business (e.g., managers, promoters, publicists,) whose main job is to build, or sometimes rebuild, a celebrity’s posi- tive image.

Finally, belief perseverance was shown to persist even when individuals were asked to consider opposite information, which in this case was to think of the reasons why others would not like a favorite celebrity. Only when the source was family or friends did the strategy of consider-the-opposite influence beliefs. However, contrary to predictions, participants were more likely to not believe the information when using this strategy. A practical implica- tion of this finding is that it shows the skepti- cism that individuals have in regard to weighing information from people who may be “haters” of the celebrity. Reducing belief perseverance has been shown in other studies of judgments and person perception; however, using the con- sider-the-opposite strategy may not be applica- ble to favored media figures or celebrities be- cause of the emotional intensity individuals have for these figures. Therefore, other tactics need to be considered to reduce perseverance of beliefs and feelings.

Future Directions

Given the exploratory nature of the present studies, future studies should replicate the pres- ent study to determine whether attitudes and beliefs would vary based on the type of celeb- rity, the specific negative (or positive) informa- tion about the celebrity, and the specific source of information (i.e., which family members, which friends or acquaintances). The present study did not compare attitudes based on the type of celebrity; however, it may benefit future researchers to evaluate whether male or female celebrities are thought of differently, or whether

46 BUI

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

ethnicity, age, or other factors of the celebrity could relate to fan attitudes. Future research can also explore whether changes in beliefs con- tinue over time. For example, Chung, Fink, and Kaplowitz (2008) tested whether message dis- crepancy and source credibility impacted be- liefs over time. They found that beliefs change at a steady and decreasing rate, but that the effect of source credibility on belief change increased over time. Chung et al.’s study did not focus on beliefs toward celebrities; however, future research could examine whether belief perseverance follows a steady but decreasing rate, as Chung et al. suggest, or whether changes are instantaneous and increase over time. This would be important in understanding how long negative beliefs about celebrities per- sist.

In the present study, participants were asked to think of the number of reasons others may have for disliking the favorite celebrity. How- ever, studies have found that when it is difficult to generate alternative explanations, belief per- severance can actually increase (Nestler, 2010). Specifically, the difficulty in accessing alterna- tive explanations convinces people that the ini- tial belief is true because if the alternatives were true they would be easier to access or think of, instead of being difficult (Nestler, 2010). Rele- vant to the present study, if there are few con- trasting or alternative information about a ce- lebrity available, it would be difficult for fans to change their emotional beliefs, even when they are asked to consider the opposite notion that others may not like the celebrity. Furthermore, future studies can investigate whether belief perseverance could be diminished in the face of continued positive information that contradicts negative media coverage of a celebrity. For instance, researchers have yet to explore whether targeted media information to change a celebrity’s image, perhaps to help the celebrity make a “comeback” or reestablish a career, would work if the information were given con- sistently and frequently. In addition, researchers can study whether explaining reasons for the negative behavior would change beliefs. Ac- cordingly, there is room for further exploration of the impact of successful and unsuccessful media campaigns to change fans’ beliefs, just as there is need to explore how to positively change beliefs about celebrities. Thus, to under- stand the power of media on fan interest, atti-

tudes, and beliefs, future research in these areas is warranted.

References

Albergotti, R., & O’Connell, V. (2013, January 18). Lance Armstrong admits to doping, ‘one big lie’. The Wall Street Journal. Retrieved from http:// online.wsj.com/article/SB1000142412788732346 8604578248582288523660.htm

Anderson, C. A., Lepper, M. R., & Ross, L. (1980). Perseverance of social theories: The role of expla- nation in the persistence of discredited. Journal of Personality and Social Psychology, 39, 1037– 1049. doi:10.1037/h0077720

Boon, S. D., & Lomore, C. D. (2001). Admirer- celebrity relationships among young adults: Explaining perceptions of celebrity influence on identity. Human Communication Research, 27, 432– 465. doi:10.1093/hcr/27.3.432

Branch, S. E., Wilson, K. M., & Agnew, C. R. (2013). Committed to Oprah, Homer, and House: Using the investment model to understand paraso- cial relationships. Psychology of Popular Media Culture, 2, 96 –109. doi:10.1037/a0030938

Chung, S., Fink, E. L., & Kaplowitz, S. A. (2008). The comparative statics and dynamics of beliefs: The effect of message discrepancy and source credibility. Communication Monographs, 75, 158 –189. doi:10.1080/03637750802082060

De Backer, C. J. S. (2012). Blinded by starlight: An evolutionary framework for studying celebrity cul- ture and fandom. Review of General Psychology, 16, 144 –151. doi:10.1037/a0027909

Eagly, A. H., Wood, W., & Chaiken, S. (1978). Causal inferences about communicators and their effect on opinion change. Journal of Personality and Social Psychology, 36, 424 – 435. doi:10.1037/ 0022-3514.36.4.424

Fiske, S. T. (2010). Ordinary personology: Figuring out why people do what they do. In Social beings (pp. 86 – 87). Hoboken, NJ: Wiley.

Fong, C. P. S., & Wyer, R. S., Jr. (2012). Consumers’ reactions to a celebrity endorser scandal. Psychol- ogy and Marketing, 29, 885– 896. doi:10.1002/mar .20571

Giles, D. C. (2000). Illusions of immortality: A psy- chology of fame and celebrity. London: Mac- Millan.

Giles, D. C., & Maltby, J. (2004). The role of media figures in adolescent development: Relations be- tween autonomy, attachment, and interest in celeb- rities. Personality and Individual Differences, 36, 813– 822. doi:10.1016/S0191-8869(03)00154-5

Guenther, C. L., & Alicke, M. D. (2008). Self- enhancement and belief perseverance. Journal of Experimental Social Psychology, 44, 706 –712. doi:10.1016/j.jesp.2007.04.010

47BELIEF PERSEVERANCE AND CELEBRITIES

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.

Jelalian, E., & Miller, A. G. (1984). The perseverance of beliefs: Conceptual perspectives and research developments. Journal of Social and Clinical Psy- chology, 2, 25–56. doi:10.1521/jscp.1984.2.1.25

Kumkale, G. T., Albarracin, D., & Seignourel, P. J. (2010). The effects of source credibility in the presence or absence of prior attitudes: Implications for the design of persuasive communication. Jour- nal of Applied Social Psychology, 40, 1325–1356. doi:10.1111/j.1559-1816.2010.00620.x

Lance’s story (n.d.). In LiveSTRONG: Who we are. Retrieved from http://www.livestrong.org/lances- story

Lepper, M. R., Ross, L., & Lau, R. R. (1986). Per- sistence of inaccurate beliefs about the self: Per- severance effects in the classroom. Journal of Per- sonality and Social Psychology, 50, 482– 491. doi: 10.1037/0022-3514.50.3.482

Lopresti, M. (2013, January 18). Lance Armstrong cancer doctor has no mixed feelings. USA Today. Retrieved from http://www.usatoday.com/story/ sports/columnist/lopresti/2013/01/17/lance- armstrong-einhorn-cancer-doctor-oprah/1842777/

Lord, C. G., Lepper, M. R., & Preston, E. (1984). Considering the opposite: A corrective strategy for social judgment. Journal of Personality and Social Psychology, 47, 1231–1243. doi:10.1037/0022- 3514.47.6.1231

Maltby, J., Day, L., McCutcheon, L. E., Houran, J., & Ashe, D. (2006). Extreme celebrity worship, fan- tasy proneness and dissociation: Developing the measurement and understanding of celebrity wor- ship within a clinical personality context. Person- ality and Individual Differences, 40, 273–283. doi: 10.1016/j.paid.2005.07.004

Maltby, J., McCutcheon, L. E., Ashe, D. D., & Houran, J. (2001). The self-reported psychological well-being of celebrity worshippers. North Amer- ican Journal of Psychology, 3, 441– 452.

McCutcheon, L. E., Lange, R., & Houran, J. (2002). Conceptualization and measurement of celebrity worship. British Journal of Psychology, 93, 67– 87. doi:10.1348/000712602162454

Mercer, J. (2010). Emotional beliefs. International Orga- nization, 64, 1–31. doi:10%10170S0020818309990221

Nestler, S. (2010). Belief perseverance: The role of accessible content and accessibility experiences. Social Psychology, 41, 35– 41. doi:10.1027/1864- 9335/a000006

Petty, R. E., & Cacioppo, J. T. (1986). Communica- tion and persuasion: Central and peripheral routes to persuasion. New York: Springer-Verlag. doi:10.1007/978-1-4612-4964-1

Ross, L., Lepper, M. R., & Hubbard, M. (1975). Perseverance in self-perception and social percep- tion: Bias attributional processes in the debriefing paradigm. Journal of Personality and Social Psy- chology, 32, 880 – 892. doi:10.1037/0022-3514.32 .5.880

Smith, K. C., Twum, D., & Geilen, A. C. (2009). Media coverage of celebrity DUIs: Teachable mo- ments or problematic social modeling? Alcohol and Alcoholism, 44, 256 –260. doi:10.1093/alcalc/ agp006

Smith, M. J. (1982). Cognitive schema theory and the perseverance and attenuation of unwarranted em- pirical beliefs. Communication Monographs, 49, 115–126. doi:10.1080/03637758209376075

Tauber, M., McNeil, L., Messer, L., Nahas, A., Lee, K., Shelasky, A., & Dennis, A. (2010, October 11). Lindsay Lohan’s road to ruin. People, 74, 52–57.

Thwaites, D., Lowe, B., Monkhouse, L. L., & Barnes, B. R. (2012). The impact of negative publicity on celebrity ad endorsements. Psychology and Mar- keting, 29, 663– 673. doi:10.1002/mar.20552

Till, B. D., & Shimp, T. A. (1998). Endorsers in advertising: The case of negative celebrity infor- mation. Journal of Advertising, 27, 67– 82. doi: 10.1080/00913367.1998.10673543

Wegner, D. M., Coulton, G. F., & Wenzlaff, R. (1985). The transparency of denial: Briefing in the debriefing paradigm. Journal of Personality and Social Psychology, 49, 338 –346. doi:10.1037/ 0022-3514.49.2.338

Werde, B. (2005, May 19). Mariah storms back. Rolling Stone, 974, 22.

Received September 25, 2012 Revision received August 13, 2013

Accepted August 20, 2013 !

48 BUI

T hi

s do

cu m

en t

is co

py ri

gh te

d by

th e

A m

er ic

an P

sy ch

ol og

ic al

A ss

oc ia

ti on

or on

e of

it s

al li

ed pu

bl is

he rs

. T

hi s

ar ti

cl e

is in

te nd

ed so

le ly

fo r

th e

pe rs

on al

us e

of th

e in

di vi

du al

us er

an d

is no

t to

be di

ss em

in at

ed br

oa dl

y.