Content uploaded by Lynn McCutcheon
Author content
All content in this area was uploaded by Lynn McCutcheon on Mar 26, 2018
Content may be subject to copyright.
A Clinical Interpretation of Attitudes and Behaviors Associated with
Celebrity Worship
JOHN MALTBY, PH.D.,
1
JAMES HOURAN, M.A.
2,3
and LYNN E. MCCUTCHEON, ED.D.
4
The phenomenon of celebrity worship is currently conceptualized as an abnormal
type of parasocial relationship, driven by absorption and addictive elements and
which potentially has significant clinical sequelae. The authors hypothesize that the
three increasingly extreme sets of attitudes and behaviors associated with celebrity
worship also partly reflect the three domains of personality discussed in Eysenckian
theory. Specifically, celebrity worship for entertainment-social reasons may reflect
extraversion personality traits; intense-personal attitudes and behaviors toward
celebrities may reflect neuroticism traits; and celebrity worship of a borderline-
pathological nature may reflect psychoticism traits. To test this idea, the authors
administered the Celebrity Attitude Scale and the Abbreviated Form of the Revised
Eysenck Personality Questionnaire to large convenience samples of students (N⫽
317) and community (N⫽290) respondents. Results indicate that celebrity worship
is not an uncommon phenomenon. Further, correlational analyses supported pre-
dictions and suggest that Eysenckian domains of personality may promote or hinder
a person’s progression along the continuum of behaviors associated with celebrity
worship.
—J Nerv Ment Dis 191:25–29, 2003
The adoration of celebrities as idols or models is
a normal part of identity development in childhood
and adolescence (Greene and Adams-Price, 1990;
Raviv et al., 1996; Yue and Cheung, 2000), but be-
yond this form of parasocial interaction is the seem-
ingly abnormal phenomenon whereby persons with
assumed intact identities become virtually obsessed
with one or more celebrities—similar to an eroto-
manic type of delusional disorder. This type of ob-
sessive-like behavior is known as celebrity worship.
Prevalence rates are not known, but celebrity wor-
ship is at least visible enough that the popular media
have taken notice. The stalking of celebrities is per-
haps the most dramatic and widely reported expres-
sion of this phenomenon (for a review see e.g.,
Melton, 2000), but celebrity worship can also affect
on a private level the person with the fixation. For
instance, in a recent issue of a teen fashion magazine
a 16-year-old girl told of her self-described obsession
with a musician and her reaction to the news of the
musician’s marriage engagement. According to
Haynes and Rich (2002), the adolescent was hospi-
talized because in response to hearing this informa-
tion she reportedly ran a hot bath and cut herself on
her neck, arms, and legs. Thoughts in her mind
during this disturbing event included, “She’s going to
change him if he gets married. . .I’m not going to live
with that” (p 198). Even on recovery from her inju-
ries, there was continued evidence of obsessional-
like ideations: “I don’t care who he dates, sleeps
with, marries. . .I realized I wanted him to be happy,
and that that would make me happy. . .He’s the only
person I connect with” (Haynes and Rich, 2002, p 198).
McCutcheon et al. (2002) proposed an Absorption-
Addiction model to explain such cases of celebrity
worship. According to this model, a compromised
identity structure in some persons facilitates psy-
chological absorption with a celebrity in an attempt
to establish an identity and a sense of fulfillment.
The dynamics of the motivational forces driving this
absorption may in turn take on an addictive compo-
nent, leading to more extreme (and perhaps delu-
sional) behaviors to sustain the person’s satisfaction
with the parasocial relationship. Several studies
based on the Celebrity Attitude Scale (Maltby et al.,
2001; Maltby et al., 2002; McCutcheon et al., 2002)
are consistent with this proposed model and suggest
that there are three increasingly more extreme sets
of attitudes and behaviors associated with celebrity
worship.
1
School of Psychology, University of Leicester, Leicester,
United Kingdom.
2
Department of Psychiatry, Southern Illinois University
School of Medicine, 901 West Jefferson, P.O. Box 19642, Spring-
field, Illinois 62794-9642. Send reprint requests to J. Houran.
3
Department of Psychology, University of Adelaide, Adelaide,
South Australia.
4
Department of General Education, DeVry University, Or-
lando, Florida.
0022-3018/03/1911–25 Vol. 191, No. 1
THE JOURNAL OF NERVOUS AND MENTAL DISEASE Printed in U.S.A.
Copyright © 2003 by Lippincott Williams & Wilkins
25
Low levels of celebrity worship have entertain-
ment-social value and comprise attitudes and behav-
iors like “My friends and I like to discuss what my
favorite celebrity has done,”and “Learning the life
story of my favorite celebrity is a lot of fun.”This
stage reflects social aspects to celebrity worship and
is consistent with Stever’s (1991) observation that
fans are attracted to a favorite celebrity because of
their perceived ability to entertain and capture our
attention. Intermediate levels of celebrity worship,
by contrast, are characterized by more intense-per-
sonal feelings, defined by items like “I consider my
favorite celebrity to be my soul mate,”and “I have
frequent thoughts about my celebrity, even when I
don’t want to.”This stage arguably reflects persons’
intensive and compulsive feelings around the celeb-
rity, akin to the obsessional tendencies of fans often
referred to in the literature (Dietz et al., 1991; Giles,
2000). The most extreme expression of celebrity
worship is labeled borderline-pathological, as exem-
plified by items like “If someone gave me several
thousand dollars to do with as I please, I would
consider spending it on a personal possession (like a
napkin or paper plate) once used by my favorite
celebrity,”and “If I were lucky enough to meet my
favorite celebrity, and he/she asked me to do some-
thing illegal as a favor I would probably do it.”This
factor is thought to reflect a person’s social-patho-
logical attitudes and behaviors that are held as a
result of worshiping a celebrity.
Sociological factors like mass media and commu-
nication may buttress all of these behaviors (e.g.,
Giles, 2000; Showalter, 1997), but a number of psy-
chological risk factors also influence a person’s pro-
gression along the continuum of celebrity worship.
In particular, we know that celebrity worshipers
exhibit poorer psychological functioning than non-
worshipers (Maltby et al., 2001; McCutcheon et al.,
in press); the phenomenon occurs more in adoles-
cents or young adults than older persons (Ashe and
McCutcheon, 2001; Giles, in press);
5
and celebrity
worshipers are more likely than nonworshipers to
value a “game-playing”love style (McCutcheon,
2002). However, celebrity worship does not appear
to be related to authoritarianism (Maltby et al., 2001)
and at best is only very weakly associated with
shyness or loneliness (Ashe and McCutcheon, 2001).
Moreover, we speculate that different dimensions
of personality in part promote susceptibility to in-
creasingly more extreme levels of celebrity worship.
For instance, the three stages discussed above
strongly parallel the three dimensions of Eysenckian
personality theory (Eysenck and Eysenck, 1985):
extraversion, neuroticism, and psychoticism. That
is, the entertainment-social factor of the Celebrity
Attitude Scale reflects some of the extraversion per-
sonality traits (sociable, lively, active, venturesome);
the intense-personal factor of the Celebrity Attitude
Scale reflects some of the neuroticism traits (tense,
emotional, moody); and some of the acts described
in the borderline-pathological subscale of the Celeb-
rity Attitude Scale seem to reflect some of the psy-
choticism traits (impulsive, antisocial, egocentric).
Therefore, the purpose of this study was to test
these predictions by examining the correlations be-
tween the three levels of celebrity worship and the
three dimensions of Eysenck’s personality theory.
Methods
Respondents
Data were collected from two samples in the
United Kingdom. The first sample consisted of 164
male and 153 female full-time university students
(mean age ⫽20.4 years, SD ⫽2.6, range ⫽18 to 29
years). The sample was predominantly composed of
single (N⫽290) white persons (N⫽211) who were
currently employed in some type of part-time work
(N⫽163). The other sample was a convenience
sample of 127 male and 163 female respondents
(mean age ⫽34.3 years, SD ⫽8.1, range ⫽22 to 60
years) from a number of workplaces and commu-
nity groups. This sample was predominantly com-
posed of married (N⫽144) white persons (N⫽
208) who were employed (N⫽197), and most
(N⫽87) had the equivalent of at least one “O”
level/GCSE education.
Materials
Materials included the a) Celebrity Attitude Scale
(CAS; McCutcheon et al., 2002). This is a 34-item
Likert-type scale with “strongly agree”equal to 5 and
“strongly disagree”equal to 1. From analysis reported
in Maltby et al. (2002), three subscales were formed
from 23 of the items: entertainment-social, intense-
personal, and borderline-pathological. Additionally,
the b) Abbreviated Form of the Revised Eysenck Per-
sonality Questionnaire (EPQR-A; Francis et al., 1992),
which contains 6-item measures of extraversion, neu-
roticism, psychoticism, and lie scores, was used. Re-
sponses are scores on a “yes”–“no”response format.
Within Eysenckian theory, extraversion traits com-
prise sociable, carefree, optimistic traits (Eysenck and
Eysenck, 1975). A person scoring high on extraversion
is likely to have many friends, take chances, and crave
5
Giles DC, Maltby J (submitted) The role of media in adoles-
cent development: Relations between autonomy, attachment,
and interest in celebrities.
MALTBY et al.
26
excitement (Eysenck et al., 2000). Second, a person
scoring high in neuroticism is anxious, moody, and
frequently depressed, likely to suffer from psycho-
somatic disorders, preoccupied with things that may
go wrong, and is overly emotional (Eysenck and
Eysenck, 1975; Eysenck et al., 2000). Third, psy-
choticism comprises solitary, troublesome, cruel, in-
humane, aggressive, and insensitive traits (Eysenck
and Eysenck, 1975; Eysenck et al., 2000). Psychiatric
terms associated with psychoticism subsume “schiz-
oid,”“psychopathic,”and “behavioral disorders;”
however, within Eysenckian theory, these are only
correlates of extreme scores on the scale (Eysenck
et al., 2000). Last, the questionnaire also contains a
measure of lie scores, to take account of persons
producing responses that are socially desirable.
We emphasize that these three dimensions of per-
sonality are not mutually exclusive; people have
these three traits to different degrees to the extent
that Eysenck et al. (2000) suggested that sometimes
personality traits are mixed. Those authors gave the
example that a person who scores high in neuroti-
cism and extraversion will likely to be touchy, rest-
less, and easily excitable. Moreover, the scale has
been subject to exploratory and confirmatory factor
analyses that suggest the unidimensionality of the
four EPQR-A subscales of Extraversion, Neuroti-
cism, Psychoticism, and the Lie Scale (Forrest et al.,
2000). This is consistent with the popular notion of a
continuum within the general population, along
which ordinary and pathological forms of thought
and perception may be mapped (e.g., Claridge, 1990,
1997; Posey and Losch, 1983–1984; Prentky, 1989).
Construct validity has been found for the subscales
in terms of predicted relationships with psychological
well-being, affect, religiosity, cognitive tasks, and
sex roles (Chang, 1997; Cooper and Taylor, 1999;
Francis and Bolger, 1997; Lewis and Maltby, 1995;
Shevlin et al., 2002).
Results
Table 1 shows the mean score and standard devi-
ation for each measure. There were no statistically
significant gender differences for any of the scales,
except that in both samples, the women scored sig-
nificantly higher than the men on neuroticism. How-
ever, an alpha correction for multiple observations
indicates this effect is not robust. Table 1 gives the
Cronbach alphas for all the scales for the university
and older adult samples. The reliability statistic for
the 4-item CAS borderline-pathological subscale in
both samples was, on first analysis, very low (student
sample,
␣
⫽.55; older adult,
␣
⫽.52). On examining
item-to-total correlations, removal of the item “News
about my favorite celebrity is a pleasant break from
the harsh world”(item 32) increased the reliability
statistic to satisfactory level for each sample (Table 1).
Given that Maltby et al. (2002) have commented on the
unusualness of this item in the borderline-pathological
scale, it was removed, and scores for a 3-item measure
of CAS borderline-pathological subscale were com-
puted. Therefore, present findings suggest that all the
scales show satisfactory internal reliability, with the
exception of psychoticism, which falls slightly below
Kline’s (1986) criterion of .70.
Contrary to the idea that celebrity worship is an
uncommon phenomenon, Table 2 shows that ap-
proximately 36% of our combined sample scored at
or above the theoretical midpoints on the three sub-
scales of the CAS. Interestingly, about 27% of the
combined sample scored highly on the intense-per-
sonal and borderline-pathological subscales. Our
sampling procedure does not allow us to generalize,
but these findings suggest the possibility that many
persons do not engage in celebrity worship for mere
entertainment. Rather, there appears to be a clear
clinical component to attitudes and behaviors asso-
ciated with celebrity worshiping.
TABLE 1
Mean Scores (SD) by Sex and Alpha Coefficients of All the Scales by Both Samples
Students (n⫽317) Older Adults (n⫽290)
␣
Males Females t(315)
␣
Males Females t(288)
CAS-ES
a
.84 19.13 (8.2) 19.93 (9.7) ⫺.79 .90 16.73 (7.6) 16.79 (8.3) ⫺.06
CAS-IP
b
.82 22.54 (7.2) 23.26 (7.9) ⫺.84 .88 20.76 (8.0) 21.13 (7.7) ⫺.39
CAS-Pathology
c
.70 03.66 (2.4) 03.97 (2.5) ⫺1.14 .70 03.42 (2.2) 03.53 (2.3) ⫺.41
Extraversion .80 03.81 (2.4) 03.47 (2.3) 1.63 .85 03.36 (2.4) 03.12 (2.4) .85
Neuroticism .78 03.08 (2.4) 03.87 (2.4) ⫺2.64* .75 02.93 (2.5) 03.68 (2.4) ⫺2.57*
Psychoticism .68 02.75 (2.5) 02.76 (2.2) ⫺.03 .66 02.81 (2.5) 02.72 (2.4) .32
Lie scores .81 02.14 (2.4) 02.11 (2.3) .11 .72 01.96 (2.2) 01.98 (2.3) ⫺.08
a
CAS-ES: Celebrity Attitude Scale ⫺Entertainment-Social.
b
CAS-IP: Celebrity Attitude Scale ⫺Intense Personal.
c
CAS-Pathology: Celebrity Attitude Scale ⫺Borderline-Pathological.
*p⬍.05; **p⬍.01; ***p⬍.001 (statistically significant after .05 Bonferroni correction for multiple observations).
PERSONALITY AND CELEBRITY WORSHIP 27
Table 3 gives the Pearson correlations between all
the variables for the two samples. Consistent with
predictions and replicating well across both sam-
ples, celebrity worship for entertainment-social rea-
sons shared a significantly positive relationship with
extraversion; celebrity worship for intense-personal
reasons shared a significantly positive correlation
with neuroticism; and celebrity worship reflecting
pathological tendencies shared a significantly posi-
tive relationship with psychoticism.
Discussion
The present findings suggest that, among UK sam-
ples, celebrity worship is not an uncommon phenom-
enon and its expression may be explained in part
within wider personality theory. That is, celebrity
worship for entertainment-social reasons is associ-
ated with extraversion; celebrity worship for in-
tense-personal reasons is associated with neuroti-
cism; and celebrity worship reflecting pathological
thoughts and behaviors is related to psychoticism,
all defined in Eysenckian terms. The effect sizes are
at best modest (all rvalues ⬍.31); however, they are
robust given a Bonferroni correction (.05) for mul-
tiple observations. Moreover, the findings are con-
sistent with predictions that derived from descrip-
tions of celebrity worship and Eysenck’s personality
dimensions.
Accordingly, Eysenckian personality theory pro-
vides a theoretical and empirical context for previ-
ous and future findings using the CAS and can be
seen as an important supplement to McCutcheon et
al.’s (2002) Absorption-Addiction model of celebrity
worship. In this respect, dimensions of personality
could help promote or hinder progressively more
extreme behaviors described by this model. That is,
the addictive component to McCutcheon et al.’s
(2002) Absorption-Addiction model may be medi-
ated partly by personality traits. For instance, we
would not expect persons who score high on the
measure of extraversion and score low on the mea-
sures of neuroticism and psychoticism to be prone
to exhibit dysfunctional expressions of celebrity
worship. By contrast, previous studies have found
celebrity worship for intense personal reasons to be
related to depression and anxiety (Maltby et al.,
2001). The present findings suggest that this aspect
of celebrity worship is positively associated with
neuroticism, and since neuroticism is clearly related
to anxiety and depression (Eysenck and Eysenck,
1975; Gaynes et al., 1997; Maltby et al., 1998), the
neuroticism factor provides a useful understanding
of why these higher levels of celebrity worship are
related to poorer mental health. It may also be the
case that Eysenckian personality dimensions regu-
late the degree to which celebrity worshipers are
uninhibited in behaviors related to their fixation.
Additionally, the findings suggest that future re-
search may find some use in using Eysenck’s per-
sonality theory as a basis to interpret and conceptu-
alize correlates of celebrity worship. The celebrity
TABLE 3
Pearson Correlations Between All the Scales
Students Above the Diagonal (n⫽317)
CAS-ES CAS-IP CAS-Path E N P L
CAS-ES
a
1.00 .164** .181** .300** .101 .033 ⫺.037
CAS-IP
b
.153** 1.00 .131** ⫺.059 .306*** .082 .002
CAS-Pathology
c
.193** .215** 1.00 .018 ⫺.002 .171** .016
Extraversion .297*** ⫺.011 .076 1.00 ⫺.033 .047 ⫺.085
Neuroticism .098 .267*** .024 .079 1.00 .117* .012
Psychoticism ⫺.010 .046 .208*** ⫺.027 .086 1.00 ⫺.060
Lie scores ⫺.041 .070 .061 ⫺.046 .009 ⫺.051 1.00
Adults Below the Diagonal (n⫽290)
a
CAS-ES: Celebrity Attitude Scale ⫺Entertainment-Social.
b
CAS-IP: Celebrity Attitude Scale ⫺Intense Personal
c
CAS-Pathology: Celebrity Attitude Scale ⫺Borderline-Pathological.
*p⬍.05; **p⬍.01; ***p⬍.001 (statistically significant after .05 Bonferroni correction for multiple observations).
TABLE 2
Distribution of Respondents who Scored at or above
Theoretical Midpoints on the Three Subscales of the Celebrity
Attitude Scale
Adult Sample (n⫽290)
Subscale Total Sample Men Women
Entertainment-Social 18 7 11
Intense-Personal 61 26 35
Borderline-Pathological 5 2 3
Student Sample (n⫽317)
Subscale Total Sample Men Women
Entertainment-Social 36 16 20
Intense-Personal 88 42 46
Borderline-Pathological 8 4 4
Entertainment-Social [10 items] ⫽30; Intense-Personal [9 items] ⫽
27; Borderline-Pathological [3 items] ⫽3
MALTBY et al.
28
worship borderline-pathological subscale was posi-
tively related to psychoticism, and psychoticism has
been thought to be important to distinguishing be-
tween tender-minded and tough-minded attitudes
(Eysenck, 1975). Therefore, future research may ex-
amine the relationship between tough-mindedness,
as measured by Hartmann’s (1991) Boundary Ques-
tionnaire or Lange et al.’s (2000) Revised Translimi-
nality Scale, and borderline-pathological celebrity
worship to extend the understanding of this most
extreme level of celebrity worship. We predict that
significant ego-boundary deficits are operating at
this stage of celebrity worship. Accordingly, the Ab-
sorption-Addiction model augmented by Eysenck’s
personality theory implies that at this stage the ex-
pression of celebrity worship is firmly rooted in
pathology and thus may become a serious clinical
issue. The case of the female adolescent who was
willing to self-mutilate over her parasocial relation-
ship with a popular musician subsequently might
serve as an illustrative example.
References
Ashe DD, McCutcheon LE (2001) Shyness, loneliness, and atti-
tude toward celebrities. Curr Res Soc Psychol 6:124 –133. [on-
line http://www.uiowa.edu/|P5grpproc/crisp.]
Chang EC (1997) Positive and negative affectivity for academic
and interpersonal domains: Relations to general affectivity,
extraversion, and neuroticism. Pers Individ Dif 22:929 –932.
Claridge G (1990) Can a disease model of schizophrenia survive?
In RP Bentall (Ed), Reconstructing schizophrenia (pp 157–
183). London: Routledge.
Claridge G (1997) Theoretical background and issues. In G
Claridge (Ed), Schizotypy: Implications for illness and health
(pp 3–18). Oxford: Oxford University Press.
Cooper C, Taylor R (1999) Personality and performance on a
frustrating cognitive task. Percept Mot Skills 88:1384.
Dietz PE, Matthews DB, Van Duyne C, Martell DA, Parry CD,
Stewart T, Warren J, Crowder JD (1991) Threatening and
otherwise inappropriate letters to Hollywood celebrities. J
Forensic Sci 36:185–209.
Eysenck HJ (1975) The structure of social attitudes. Br J Soc
Clin Psychol 14:323–331.
Eysenck HJ, Eysenck MW (1985) Personality and individual
differences: A natural science approach. New York: Plenum.
Eysenck HJ, Eysenck SBG (1975) Manual of the Eysenck Per-
sonality Questionnaire. London: Hodder & Stoughton.
Eysenck HJ, Wilson GD, Jackson CJ (2000) Manual of the
Eysenck Personality Profiler, (Vol 6). Guildford, UK: PSi-
Press.
Forrest S, Lewis CA, Shevlin M (2000) Examining the factor
structure and differential functioning of the Eysenck person-
ality questionnaire revised—abbreviated. Pers Individ Dif 29:
579 –588.
Francis LJ, Bolger J (1997) Personality and psychological well-
being in later life. Irish J Psychol 18:444 –447.
Francis LJ, Brown LB, Philipchalk R (1992) The development of
an Abbreviated Form of the Revised Eysenck Personality
Questionnaire (Epqr-A)—Its use among students in England,
Canada, the USA and Australia. Pers Individ Dif 13:443–449.
Gaynes BN, Magruder KM, Burns BJ, Broadhead WE (1997)
Persistence of depressive illness in primary care patients with
major depression: Is a coexisting anxiety disorder a risk fac-
tor? Int J Psychiatry Med 27:307–308.
Giles D (2000) Illusions of immortality: A psychology of fame
and celebrity. London: MacMillan.
Giles DC (in press) Parasocial interaction: A review of the liter-
ature and a model for future research. Media psychology.
Greene AL, Adams-Price C (1990) Adolescents’secondary attach-
ment to celebrity figures. Sex Roles 23:335–347.
Hartmann E (1991) Boundaries in the mind: A new psychology
of personality. New York: Basic Books.
Haynes E, Rich N (2002) Obsessed fans! YM [Your Magazine]
50:196 –199.
Kline P (1986) A handbook of test construction. London:
Methuen.
Lange R, Thalbourne, MA, Houran J, Storm L (2000) The Revised
Transliminality Scale: Reliability and validity data from a
Rasch top-down purification procedure. Conscious Cogn
9:591–617.
Lewis CA, Maltby J (1995) Religiosity and personality among US
adults. Pers Individ Dif 18:293–295.
Maltby J, Houran J, Lange R, Ashe D, McCutcheon LE (2002)
Thou shalt worship no other gods—unless they are celebrities:
The relationship between celebrity worship and religious ori-
entation. Pers Individ Dif 32:1157–1172.
Maltby J, Lewis CA, Hill AP (1998) Oral pessimism and depres-
sive symptoms: A comparison with other correlates of depres-
sion. Br J Med Psychol 71:195–200.
Maltby J, McCutcheon LE, Ashe DD, Houran J (2001) The self-
reported psychological well-being of celebrity worshippers. N
Am J Psychol 3:441–452.
McCutcheon LE (2002) Are parasocial relationship styles re-
flected in love styles? Curr Res Soc Psychol 7:82–94. [electron-
ic journal]. Available: http://www.uiowa.edu/|P5grpproc/crisp/
crisp.7.6.htm.
McCutcheon LE, Ashe DD, Houran J, Maltby J (in press) A
cognitive profile of individuals who tend to worship celebri-
ties. J of Psychol.
McCutcheon LE, Lange R, Houran J (2002) Conceptualization
and measurement of celebrity worship. Br J Psychol 93:67–87.
Melton HC (2000) Stalking: A review of the literature and direc-
tion for future research. Criminal Justice Rev 25:246 –262.
Posey TB, Losch ME (1983–1984) Auditory hallucinations of
hearing voices in 375 normal subjects. Imagination Cogn Pers
3:99 –113.
Prentky R (1989) Creativity and psychopathology: Gambling at
the seat of madness. In JA Glover, RR Ronning, CR Reynolds
(Eds), Handbook of creativity (pp 243–269). New York: Ple-
num.
Raviv A, Bar-Tal D, Raviv A, Ben-Horin A (1996) Adolescent
idolization of pop singers: Causes, expressions, and reliance. J
Youth Adolesc 25:631–650.
Shevlin M, Bailey F, Adamson G (2002) Examining the factor
structure and sources of differential functioning of the
Eysenck Personality Questionnaire Revised—Abbreviated.
Pers Individ Dif 32:479 –487.
Showalter E (1997) Hystories: Hysterical epidemics and mod-
ern media. New York: Columbia University Press.
Stever GS (1991) The Celebrity Appeal Questionnaire. Psychol
Rep 68:859 –866.
Yue XD, Cheung C-K (2000) Selection of favourite idols and
models among Chinese young people: A comparative study in
Hong Kong and Nanjing. Int J Behav Dev 24:91–98.
PERSONALITY AND CELEBRITY WORSHIP 29
