Celebrity Worship Syndrome

Celebrity worship syndrome (CWS) is an obsessive-addictive disorder in which a person becomes overly involved with the details of a celebrity's personal life.

Objects of affection
Psychologists have indicated that though many people obsess over glamorous film, television, sport and pop stars, others have unlikely icons such as politicians or authors. The only common factor between them is that they are all figures in the public eye (i.e., celebrities). The term Celebrity Worship Syndrome is in fact a misnomer.

The term celebrity worship syndrome (CWS) first appeared in an article 'Do you worship the celebs?' by James Chapman in the Daily Mail in 2003 (Chapman, 2003). James Chapman was basing his article on the journal paper Maltby et al. (2003). James Chapman refers to CWS, but in fact this is a misunderstanding of a term used in the academic article to which he refers (Maltby et al. 2003), CWS which stood for Celebrity Worship Scale. Nonetheless Chapman may be generally correct. A syndrome refers to a set of abnormal or unusual set of symptoms indicating the existence of an undesirable condition or quality. Indeed many attitudes and behaviours covered in this research indicate such states.

Three types of CWS
Psychologists in the U.S.A. and UK. created a celebrity worship scale to rate the problems. In 2002, United States psychologists Lynn McCutcheon, Rense Lange, and James Houran introduced the Celebrity Attitude Scale, a 34 item scale administered to 262 persons living in central Florida. McCutcheon et al. suggested that celebrity worship comprised one dimension in which lower scores on the scale involved individualistic behavior such as watching, listening to, reading and learning about celebrities whilst the higher levels of worship are characterized by empathy, over-identification, and obsession with the celebrity.

However, later research among larger UK samples have suggested there are 3 different aspects to celebrity worship; John Maltby (University of Leicester), and the aforementioned psychologists examined the Celebrity Attitude Scale among 1732 United Kingdom respondents (781 males, 942 females) who were aged between 14 and 62 years and found the following 3 dimensions to celebrity worship:

Entertainment-social
This dimension comprises attitudes that fans are attracted to a favorite celebrity because of their perceived ability to entertain and become a social focus such as “I love to talk with others who admire my favorite celebrity” and “I like watching and hearing about my favorite celebrity when I am with a large group of people”.

Intense-personal
Intense-personal aspect of celebrity worship reflects intensive and compulsive feelings about the celebrity, akin to the obsessional tendencies of fans often referred to in the literature; for example “I share with my favorite celebrity a special bond that cannot be described in words” and “When something bad happens to my favorite celebrity I feel like it happened to me’”.

Borderline-pathological
This dimension is typified by uncontrollable behaviors and fantasies regarding scenarios involving their celebrities, such as “I have frequent thoughts about my favorite celebrity, even when I don’t want to” and “my favorite celebrity would immediately come to my rescue if I needed help”.

Celebrity worship and mental health
Evidence indicates that poor mental health is correlated with celebrity worship. Researchers have examined the relationship between celebrity worship and mental health in United Kingdom adult samples. Maltby et al. (2001) found evidence to suggest that the intense-personal celebrity worship dimension was related to higher levels of depression and anxiety. Similarly, Maltby et al., in 2004, found that the intense-personal celebrity worship dimension was not only related to higher levels of depression and anxiety, but also higher levels of stress, negative affect, and reports of illness. Both these studies showed no evidence for a significant relationship between either the entertainment-social or the borderline-pathological dimensions of celebrity worship and mental health.

Another correlated pathology was recently reported by Maltby, Giles, Barber and McCutcheon (2005) who examined the role of celebrity interest in shaping body image cognitions. Among three separate UK samples (adolescents, students and older adults) individuals selected a celebrity of their own sex whose body/figure they liked and admired, and then completed the Celebrity Attitude Scale along with two measures of body image. Significant relationships were found between attitudes toward celebrities and body image among female adolescents only. The findings suggested that, in female adolescence, there is an interaction between intense-personal celebrity worship and body image between the ages of 14 and 16 years, and some tentative evidence is found to suggest that this relationship disappears at the onset of adulthood, 17 to 20 years. These results are consistent with those authors who stress the importance of the formation of relationships with media figures, and suggest that relationships with celebrities perceived as having a good body shape may lead to a poor body image in female adolescents.

Within a clinical context the effect of celebrity might be more extreme, particularly when considering extreme aspects of celebrity worship. Maltby, Day, McCutcheon, Houran and Ashe (2006) examined the relationship between entertainment-social, intense-personal and borderline-pathological celebrity worship and obsessiveness, ego-identity, fantasy proneness and dissociation. Two of these variables drew particular attention; fantasy proneness (time spent fantasising, reporting hallucinatory intensities as real, reporting vivid childhood memories, having intense religious and paranormal experiences) and dissociation (reflects the lack of a normal integration of experiences, feelings, and thoughts in everyday consciousness and memory and is related to a number of psychiatric problems).

Though low levels of celebrity worship (entertainment-social) are not associated with any of the clinical measures, medium levels of celebrity worship (intense-personal) are related to fantasy proneness (around 10% of the shared variance), while high levels of celebrity worship (borderline-pathological) share a greater association with fantasy proneness (around 14% of the shared variance) and dissociation (around 3% of the shared variance, though the effect size of this is small and most probably due to the large sample size). This finding suggests that as celebrity worship becomes more intense, and the individual perceives having a relationship with the celebrity, the more the individual is prone to fantasies.

Critical reflection on celebrity worship and mental health
"Celebrity worship" is a term coined by Lynn E. McCutcheon (DeVry University), Diane D. Ashe (Valencia Community College), James Houran (Southern Illinois University), John Maltby (University of Leicester) and a few further collaborators in a serious of articles published primarily in the minor (academically irrelevant) North American Journal of Psychology and a non-peer reviewed working paper serious called Current Issues in Social Psychology, but also in the (previously) highly respectable Journal of Psychology and British Journal of Psychology. However, as the findings of McCutcheon, Maltby, Houran, Ashe & Co's publications are increasingly presented, cited and uncritically taken at face value in the popular media, despite the actual absence of evidence (Their findings are in fact contradicted by their own published data in each of their studies!), a critical reflection is urgently required. This becomes particularly necessary, as a number of historical (Barbas 2001; Hansen 1991), ethnographic (i.e. Henry & Caldwell 2007; Jenkins 1992; Kozinets 2001; O'Guinn 1991; Richardson & Turley 2006; Stacey 1994); netnographic (i.e. Kozinets 1997) and autoethnographic studies (i.e. Holbrook 1987, 1995; Wohlfeil and Whelan 2008) in diverse academic disciplines such as film studies, media studies, cultural studies and consumer research, which - unlike McCutcheon et al. or Maltby et al., who focused mainly on a student sample (with 2 exceptions) - have actually studied real fans in the field, have come to very different conclusions that are more in line with Horton & Wohl's (1956) original concept of parasocial interaction or an earlier study by Leets et al. (1995). But even scientifically, the Celebrity Worship Syndrome stands on extremely weak conceptual footings, as the studies on which their arguments are based are seriously flawed.

The term "celebrity worshipper" is essentially McCutcheon et al.'s official and rather derogatory label for fans of celebrities in general, whereby they refer indiscriminately to anybody, who has in some form an interest in a particular celebrity or celebrity culture in general. Whether an individual likes the artistic performances of a certain musician, band, actor or athlete in private or whether the individual is an enthusiastic follower of celebrity culture or whether the individual is indeed a stalker makes no difference to McCutcheon et al.'s conceptualisation of fans (McCutcheon et al. 2002, 'Conceptualization and Measurement of Celebrity Worship', British Journal of Psychology, 93 (1), 67-87). In McCutcheon et al.'s (2002, 2003, 2006) view, they are all the same. In short, their description of celebrity worshippers applies to everyone, who claims to be a fan of a certain celebrity (like most teenagers, but also adults do), who buys/reads celebrity gossip magazines (given the global sales figures, a lot of people!) and/or autobiographies (dito) and who enjoys cultural products such as music, films, theatre, sports events, books, etc. because of the respective musicians, actors, athletes, authors, etc.

Moreover, McCutcheon et al. argue - based on the supposed findings of their studies (Their own data actually contradicts the findings.) - that celebrity worshippers are 'mentally different' (aka abnormal) from so-called 'normal' people (Though they never clearly define what they consider to be 'normal' in the first place), because they are of lesser intelligence, 'cognitively inflexible' (=> stupid), dull, gullible and lack imagination (i.e. McCutcheon et al. 2003, 'A Cognitive Profile of Individuals Who Tend to Worship Celebrities', Journal of Psychology, 137 (4), 309-322). Indeed, McCutcheon et al. sought to prove in their Journal of Psychology article that fans of celebrities display a general deficit in their '(verbal) creativity, crystallized intelligence, critical thinking, spatial ability, arithmetic skills and need for cognition' (=> problem solving), which in their opinion makes fans inherently prone to worshipping celebrities. McCutcheon et al. proposed therefore that "celebrity worshipping" would constitute a mental illness that may even be genetic (Maltby et al. 2004, 'Personality and Coping: A Context for Examining Celebrity Worship and Mental Health', British Journal of Psychology, 95, 411-428).

Now, the major problem with their arguments and research findings is that, despite all their claims for the supposedly strong evidence in their research data, not a single finding in any of their studies is actually supported by their own statistical data published in the articles. Quite the opposite is the case. In fact, the interpretation errors and the flawed research design, on which they are based, are so bluntly obvious that anyone with a basic knowledge of statistical analysis (that every quantitative-oriented academic researcher should normally have) could spot them quite easily. For example, the Journal of Psychology article alone contains no less than 28 major flaws/errors with regard to methodology, data analysis and interpretation; each of them serious enough to question the rigour and findings of the entire study. The other articles published in British Journal of Psychology, North American Journal of Psychology, etc., unfortunately, aren't of any better quality either (which somehow questions the academic rigour and methodological competence of the respective reviewers and editors that passed the articles).

For example, it seems that McCutcheon et al. (2003) either didn't know what statistical significance (p-value) and practical significance (R and respectively R square) actually mean and where the difference is (i.e. Hair et al.'s (2006) "Multivariate Data Analysis" would provide some good explanations) or ignored the results deliberately, because they contradicted the researchers' strongly defended a priori assumptions. Fact is that in the Journal of Psychology study, despite claiming a strong support for the relationship between all six predicting variables and celebrity worship as depending variable, 2 of the predicting variables (arithmetic skills, need for cognition) were actually statistically insignificant in the bivariate regression analyses - meaning there's no difference between fans and 'normal' people. With regression coefficients (R values) ranging between .3 and .4, the other four predicting variables provide hardly strong support, as each of them only explains 9-16% of the variance in the celebrity worship variable. In case of the multiple regression analyses, none of predicting variables yielded a statistically significant beta value, which McCutcheon et al. choose conveniently to ignore in their findings and followed up with an interpretation that 'flies in the face of academic rigour in quantitative research' (Hair et al. 2006).

Equally important is to note that the celebrity attitude scale (CWS) is not the reliable measurement for celebrity worship, as McCutcheon et al. suggests. In fact, contradicting their initial claim that the three subcategories 'entertainment-social', 'intense-personal' and 'borderline-pathological' emerged from a factor analysis of the data, the CWS scale was actually from the start set up that way in order to extract in form of a self-fulling prophecy three factors of intensity (McCutcheon et al. 2002). Moreover, the current 23-item scale came only into being because the original 34-item scale failed to work and was reduced until a reliability test of the data yielded an acceptable Cronbach Alpha (see Hair et al. 2006). Interesting is also that the ranking order (low, middle, high) of the three subcategories has regularly changed between published studies as a result of obtaining some curious scores. A reason for this regular anomaly may be the fact that many items in the CWS contain actually leading, ambiguous, emotionally-charged questions (which should be avoided in proper questionnaire design, see Hair et al. 2006). Actually, the CWS seems only to work with students in a specific US college (The authors used primarily a student sample, where participation was encouraged through course credits.), while the scale collapsed in case of samples comprising British housewives or schoolgirls (see several McCutcheon et al. papers published between 2001 and 2007 in the North American Journal of Psychology and Current Issues in Social Psychology). Yet, the latter results have never been cited or mentioned in any of their other publications nor discussed in the findings of the same papers.

In light of this series of shortcomings, the question is why those obviously flawed studies got nonetheless published in reputable journals such as the Journal of Psychology and the British Journal of Psychology, which pride themselves on ensuring quality through a rigorous academic review process. The answer lies in the politics of publication within academia that are unknown to the general public. Both journals are dominated by proponents a specific positivist research ontology that adheres faithfully to the traditional (but out-dated) neo-behaviorist research paradigm of experimental psychology, which holds that only observations by researchers and statistcal data generated in experiments can be considered 'trustworthy', reliable and valid. Verbal statements of subjects (i.e. interviews), on the other hand, are seen within this paradigm as unreliable and can't be trusted, because individuals "don't know why they are doing what they do and, therefore, are incapable of describing their respective thoughts and emotions accurately. What they provide instead are posthumous justifications of their actions." (Nisbett & Wilson 1977; a much-cited, but methodologically flawed paper). With new alternative and often interpretive/humanistic research paradigms and methodologies emerging in the social sciences since the mid-1970s (i.e. Hirschman & Holbrook 1992), the editors at top journals have become more and more concerned with securing the continuing dominance of their traditional research paradigm (see Calder & Tybout 1987 as an example) rather than focusing on the publication of interesting new ideas. An unfortunate consequence is that editors and reviewers, these day, pay much more attention to the manual-like adherence to the prescribed methodological procedures than to the substance and quality of the actual content of a study (Arnould & Thompson 2005; Calder & Tybout 1987; Hair et al. 2006; Holbrook 1995).