Celebrity Worship Among Universi

HOGREFE PUBLISHING - AUTHOR QUERY FORM

Journal Code: EPP Article Number: a000029 Article Title: Celebrity Worship Among University Students in Malaysia: A Methodological Contribution to the Celebrity Attitude Scale AUTHOR QUERIES – TO BE ANSWERED BY THE CORRESPONDING AUTHOR Dear Author, During the preparation of your manuscript for typesetting, the queries listed below have arisen. Please answer these queries by marking the required corrections at the appropriate point in the text. Electronic file usage Sometimes we are unable to process the electronic file of your article and/or artwork. If this is the case, we have proceeded by:

Scanning (parts of) your article

Rekeying (parts of) your article

Queries and/or remarks

Location

Query / remark

Page 9 (line no 814)

Please approve the edit of the sentence “Subash K. Pillai … adolescents”.


Please approve the edit of “Bonferonni” to “Bonferroni” in the sentence “To examine … variable”.


Please provide the text for footnote 1 cited in the sentence “A chi-square … p < .001.1”.


Please clarify the sentence “For example, … hedonism)”.


Please clarify the sentence “Similarly ... samples”.

Page 8 (line nos 725 and 745)

Please update the references “Swami (in press)” and “Swami et al. (in press)”.

Page 1 (line nos. 39 and 43)

Please note that the references “Turner, Marshall, & Bonner, 2000”, “Erdogan, 1995” have been changed to “Turner, Bonner, & Marshall, 2000” and “Erdogan, 1999” as per the reference list.


Please note that “Mather. I.” has been changed to “Mathur, I.” in the reference “Mathur et al. (1997)” as per Net information.

Page 2 and 7 (line nos. 136 and 512)

The references “Cheung & Yue, 1999” and McCutcheon et al. (2004) are cited in the text but not listed. Please check.

CE: Chandu Thank you for your assistance.

Response

32 1

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

Celebrity Worship Among University Students in Malaysia A Methodological Contribution to the Celebrity Attitude Scale Viren Swami,1,2 Tomas Chamorro-Premuzic,3 Khairul Mastor,4 Fatin Hazwani Siran,4 Mohammad Mohsein Mohammad Said,4 Jas Jaafar,5 Dhachayani Sinniah,6 and Subash K. Pillai7

Un - n co ot rre fo cte rd d ist pr rib oo ut fs io n

4

Original Articles and Reviews

1

Department of Psychology, University of Westminster, UK Department of Psychology, HELP University College, Malaysia 3 Department of Psychology, Goldsmiths, University of London, UK 4 Centre for General Studies, Universiti Kebangsaan Malaysia, Malaysia 5 Department of Anthropology and Sociology, Universiti Malaya, Malaysia 6 Department of Psychiatry, Universiti Kebangsaan Malaysia, Malaysia 7 Medical Centre, Universiti Malaya, Malaysia 2

Abstract. The present study examined conceptual issues surrounding celebrity worship in a Malay-speaking population. In total, 512 Malay and 269 Chinese participants from Malaysia indicated who their favorite celebrity was and completed the Celebrity Attitude Scale (CAS) as well as a range of demographic items. Results showed that the majority of Malay and Chinese participants selected pop stars and movie stars as their favourite celebrities, mirroring findings in Western settings. In addition, exploratory factor analysis revealed a three-factor solution of the CAS that was consistent with previous studies conducted in the West. Structural equation modeling further revealed that participant’s age was negatively associated with celebrity worship and that self-rated attractiveness was positively associated with celebrity worship. Overall, the present results suggest that celebrity worship in Malaysia may be driven by market and media forces, and future research may well be guided by use of the CAS. Keywords: celebrity worship, idol selection, multiethnic, cross-cultural, Malaysia

26 27 29 28 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47

In the past several decades, there has been a growing interest in the sociological and psychological study of celebrity and stardom (Johansson, 2005). Beginning in the 1970s, a number of authors examined the popular representation of celebrities (e.g., Alberoni, 1972; Dyer, 1979), culminating in Boorstin’s (1961, p. 57) description of a celebrity as ‘‘a person who is known for his well-knownness . . . a human pseudo-event.’’ More recently, media and cultural analyses of celebrities have given way to an examination of contemporary ‘‘celebrity culture’’ (e.g., Giles, 2000; Rojek, 2001; Turner, 2004; Turner, Bonner, & Marshall, 2000). Within this body of work, it is also possible to discern a focus on the effects of celebrity, particularly in terms of endorsements, on consumer behavior (e.g., Agrawal & Kamakura, 1995; Erdogan, 1999; Mathur, Mathur, & Rangan, 1997; Till & Shimp, 1998). Nor have psychologists been averse to studying celebrities themselves, with early work focusing on the negative effects of fame (e.g., Frommer, Juetteman-Lembke, Stratkoetter, & Ó 2009 Hogrefe Publishing

Tress, 1995; Schaller, 1997; see also Giles, 2000). More recently, psychologists have developed an interest in celebrity worship or appreciation, that is, a focus on the adoration of celebrities from the point of view of consumers. Although parasocial interactions, including the adoration of celebrities as role models, are viewed as a normal part of identity development (e.g., Giles & Maltby, 2004; Yue & Cheung, 2000a), such worshipping usually decreases in intensity with age (Ashe & McCutcheon, 2001; Giles, 2002; Raviv, Bar-Tal, Raviv, & Ben-Horin, 1996). In some circumstances, such adoration can evoke pathological responses, ranging from vicarious bereavement (e.g., Marsden, 1997) to erotomania (e.g., Vigano, 1996). In order to explain such celebrity worship, McCutcheon, Lange, and Houran (2002) proposed an ‘‘absorptionaddiction’’ model of celebrity worship. According to these authors, psychological absorption with a celebrity is heightened in some individuals who have a compromised identity structure (e.g., introversion or a lack of meaningful European Psychologist 2010; Vol. xx(x):xxx–xxx DOI: 10.1027/1016-9040/a000029

48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66

2

relationships). Such absorption results in a heightened sense of reality of the idolized celebrity and helps the individual to establish a stronger sense of identity and role fulfillment. Over time, however, celebrity worship takes on an addictive component in order to maintain adequate connectedness to the celebrity, and this can lead to more extreme and pathological forms of worship (McCutcheon et al., 2002). A number of recent studies have provided some support for this model (e.g., Maltby, Day, McCutcheon, Houran, & Ashe, 2006; Maltby, Houran, Lange, Ashe, & McCutcheon, 2002; Maltby et al., 2004; McCutcheon et al., 2002). Based on item reductions of the Celebrity Attitude Scale (CAS), these studies have suggested that there are three core attitudinal components of celebrity worship, and that these components become progressively more extreme. Specifically, the first component (Entertainment-social) reflects an ‘‘attraction to a favorite celebrity because of their perceived ability to entertain and social focus.’’ The second component (Intensepersonal) refers to ‘‘intensive and compulsive feelings about the celebrity’’ whereas the final component (Borderlinepathological) is the most extreme expression of celebrity worship and reflects ‘‘an individual’s social-pathological attitudes and behaviors’’ (Maltby et al., 2006, p. 274). To be sure, the CAS is not the only available measure of celebrity worship (for a review of other scales, see McCutcheon et al., 2002, pp. 69–70), but it is perhaps the most reliable and certainly the most widely examined. For instance, recent work has examined the CAS in relation to, among other factors, aspects of religiosity (Maltby et al., 2002), body image (Maltby, Giles, Barber, & McCutcheon, 2005; Swami, Taylor, & Carvalho, 2009), acceptance of cosmetic surgery (Swami, Taylor, & Carvalho, in press), coping styles (Maltby et al., 2004), boundary structure (Houran, Navik, & Zerrusen, 2005), attributional style and self-esteem (North, Sheridan, Maltby, & Gillett, 2007), fantasy proneness and dissociation (Maltby et al., 2006), and personality (Maltby, Houran, & McCutcheon, 2003; Maltby et al., 2004). In general, these studies support the three-dimensional nature of the CAS and, more generally, McCutcheon et al.’s (2002) absorption-addiction model of celebrity worship. To date, the main limitation of the CAS is that its factor structure has only been examined and validated in the United States and Britain. This is important because celebrities often echo symbolic meanings and values that are specific to a national context (McCracken, 1989; Praet, 2002). This may give rise to different forms of celebrity worship that are congruent with the national orientations or values of a society. Indeed, this has been the suggestion of Yue and colleagues (e.g., Cheung & Yue, 2000, 2003; Yue, 2000; Yue & Cheung, 2000a, 2000b), who have examined celebrity worship in mainland China and Hong Kong. Specifically, these authors distinguish between idols or celebrities on the one hand and role models on the other, and suggest that each plays a specific functional role within particular national settings. As an example, Yue and Cheung (2000a) report that the criteria for idol selection differ between participants in mainland China (where the idolization of public models has been heavily promoted by the government) and Hong Kong (where commercialism and market forces have shaped celebrity worship). More precisely, while mainland Chinese

tend to idolize models who are perceived as realistic, practical, successful, and socially altruistic, participants in Hong Kong are more likely to emphasize hedonistic, extraordinary, romantic, and commercial aspects of an idol. This is reflected in their selection of real idols: Participants in Hong Kong were more likely to idolize pop singers than their mainland Chinese counterparts, who in turn were more likely to idolize various idols promoted by the government and that show morally desirable characteristics (e.g., honesty, modesty, and dedication to society) (Cheung & Yue, 1999; Yue, 2000; Yue & Cheung, 2000a).

127 128 129 130 131 132 133 134 135 136 137

The Present Study

138

To date, however, it remains unclear to what extent similar patterns of results will be found in other non-Western settings. To overcome this dearth in the literature, the present study sought to examine patterns of celebrity selection in Malaysia, a multiethnic nation in southeast Asia. In the past decade, Malaysian has witnessed extensive economic neo-liberalism and the attendant deregulation of the mass media, resulting in a proliferation of media outlets (e.g., Abdul Wahab, 2006). In this sense, Malaysia is more akin to Hong Kong than it is to China, and could be described, in the parlance of Yue and Cheung (2000a, p. 97), as a ‘‘three-star worship’’ culture (as opposed to a ‘‘model worship’’ culture, where role models are idolized). In the first instance, therefore, we expected Malaysians to select pop stars, movie stars, and sports stars (rather than politicians or government-endorsed models) as idols, mirroring Malaysians’ admiration for glamor, wealth, and youthfulness (cf. Yue, 1999). In addition, we also conducted the first examination of the factor structure of the CAS in a non-Western setting, which could potentially highlight the extent to which the factors associated with celebrity worship are stable across cultures. Moreover, we were able to examine whether there would be significant differences in the factor structure and factor scores of the CAS among ethnic Malay and Chinese in Malaysia, which we believe is the first multiethnic test of the CAS. Although these ethnic groups have different historical, linguistic, and religious trajectories that may affect their cognitions (e.g., Hewstone & Ward, 1985), it is also important to consider that ethnic differences in psychological phenomena will likely be mitigated by the similar (or identical) cultural background shared by Malaysians, particularly among younger age groups. Indeed, recent work has shown few, if any, significant ethnic differences in various psychological phenomena among Malaysians of different ethnic groups (e.g., Swami, in press; Swami & Chamorro-Premuzic, 2008, 2009; Swami & Toveé, 2005a, 2005b; Swami et al., 2008). The present study also examined the association between self-rated attractiveness and the components of the CAS. Recent theorizing has suggested that self-rated attractiveness is conceptually different from body image and that it may influence everyday interpersonal interactions (e.g., Swami & Furnham, 2008; Swami, Furnham, Georgiades, & Pang, 2007). However, it remains unclear whether self-rated attractiveness will also influence parasocial interactions, and as a speculative test of this possibility, we included a simple

139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183


67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126

V. Swami et al.: Celebrity Worship

European Psychologist 2010; Vol. xx(x):xxx–xxx

Ó 2009 Hogrefe Publishing


measure of self-rated attractiveness in the present study. As a final extension to previous work, we conducted both exploratory factor analysis and structural equation modeling (SEM) to test overall and individual effects of participant demographics (specifically, age and sex) and self-rated attractiveness on the CAS components, while accounting for the overlap between predictors.

191

Method

192

Participants

193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225

All participants were recruited from Kuala Lumpur, the largest city in Malaysia with a population of some 2 million. As an independent federal territory, Kuala Lumpur is now the powerhouse of the Malaysian economy and had a GDP per capita of about US$8,000 (Government of Malaysia, 2001). Demographically, the city compromises many different ethnic groups (mirroring the distribution of ethnicities in Malaysia more generally), with large populations of Chinese and Malay individuals. As noted above, although there are likely ethnic differences in terms of language use, religion, and political affiliations, these do not appear to have any major effect on the expression of various psychological phenomena (e.g., Swami, in press; Swami & Chamorro-Premuzic, 2008, 2009; Swami & Toveé, 2005a, 2005b; Swami et al., 2008). That is, Malays and Chinese in Kuala Lumpur appear to inhabit a similar sociocultural context that minimizes any difference in the magnitude or direction of differences in cognition and behavior, particularly among younger age groups. The participants of this study were 781 university students enrolled in a range of different courses at two national universities in Kuala Lumpur, where the medium of teaching is in Bahasa Malaysia (Malay). Of the total sample, 512 participants (257 women and 255 men) were ethnic Malays (age M = 21.95, SD = 1.46). All participants in this group were Muslims and the vast majority were single (99.4%). The remaining 269 participants (135 women and 134 men) were ethnic Chinese and had a mean age of 21.73 years (SD = 1.40). The majority of Chinese participants were Buddhists (80.3%; Christians = 15.6%; other = 4.1%) and most were single (96.7%). All participants in this study were born and raised in Malaysia, and were (at the very least) second-generation Malaysian citizens.

226

Materials

227 228 229 230 231 232 233 234

CAS (McCutcheon et al., 2002). Originally termed the Celebrity Worship Scale, the CAS is a 34-item measure in which respondents are asked to indicate their attitude toward their favorite celebrity that they themselves have named. Among Western samples (e.g., Maltby et al., 2006), the scale has been found to have a three-factor structure comprising Entertainment-social (sample item: ‘‘Keeping up with news about my favorite celebrity is an entertaining

pastime’’), Intense-personal (sample item: ‘‘To know my favorite celebrity is to love him/her’’), and Borderlinepathological (sample item: ‘‘I have frequent thoughts about my favorite celebrity, even when I don’t want to’’). Items are rated on a 5-point scale (1 = Strongly disagree and 5 = Strongly agree). The Malay version of the CAS was developed using the standard back-translation technique (Breslin, 1970). The first author initially translated the CAS into Malay, and this version was then back-translated into English by an independent translator unaffiliated with the study. Minor differences that emerged during this process were resolved between translators, and the final version was piloted with eight Malaysian undergraduates (3 women and 5 men) in Britain who were fluent in Malay. These participants were asked to pick out any phrases that were difficult to understand, and based on their responses, minor revisions were made to the translation. Favorite celebrity. Participants were asked to name their favorite celebrity, and these data were coded inclusively according to one of six categories, namely ‘‘pop stars,’’ ‘‘movie stars,’’ ‘‘sports stars,’’ ‘‘politicians,’’ ‘‘not sure,’’ and ‘‘none.’’ Participants were also asked to state on a binary (1 = Yes and 2 = No) whether they had ever met their favorite celebrity and whether they had ever met any other celebrity. Self-rated attractiveness. Participants were asked to rate their self-assessed physical attractiveness on a 7-point Likert-type scale (1 = Not at all physically attractive and 7 = Very physically attractive). Although not all single-item scales of this type are valid and most have been criticized (Swami et al., 2007), they are nevertheless relatively robust and are reliably related to other self-perceptions (e.g., Henderson-King, Henderson-King, & Hoffman, 2001). Demographics. Participants provided their demographic details consisting of sex, age, ethnicity, religion, and marital status.

235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270

Procedure

271

Once ethical permission was obtained, all participants were tested in large lecture theaters and in the presence of examiners who answered any questions beforehand and subsequently ensured that the test was completed appropriately and in silence. The questionnaire took 15 min to complete and participants were verbally debriefed by experimenters following its completion. All participants took part on a voluntary basis and were not remunerated for their participation.

272 273 274 275 276 277 278 279 280

Results

281

Favorite Celebrity

282

Among Malay participants, 66.4% selected a pop star as their favorite celebrity, while 13.9% chose a movie star, 1.4% chose a sports star, 4.3% chose a politician, 0.8% were

283 284 285


184 185 186 187 188 189 190

3



4


not sure, and 13.3% did not have a favorite celebrity. Among Chinese participants, 35.7% chose a pop star as their favorite celebrity, 41.3% chose a movie star, 6.3% chose a sports star, 1.5% chose a politician, and 15.2% self-reported as not having any favorite celebrity. A chi-square test showed that there were significant ethnic differences in these selections, v2(5) = 107.45, p < .001.1 However, when pop, movie, and sports stars were combined, there was no longer a significant ethnic difference in these selections, v2(3) = 6.78, p = .079. In addition, 18.6% of Malays reported having met their favorite celebrity, and 81.3% reported having met some other celebrity. Among Chinese participants, 21.2% reported that they had met their favorite celebrity, and 59.1% reported that they had met some other celebrity. These data were not analyzed further.

301

Data Reduction

302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329

First, exploratory factor analysis was carried out in order to identify the underlying structure of the CAS items. It is noteworthy that, even though the current study attempted to replicate a previously reported solution, confirmatory methods (e.g., confirmatory factor analyses, CFA) are inappropriate at the item level, primarily due to the skewed distribution of data at this level (Bernstein & Teng, 1989; for a more recent discussion of this point, see Petrides, Jackson, Furnham, & Levine, 2004). We, therefore, opted for a Principal Axis Factoring for our initial data reduction. Missing values (fewer than 2.0%) were replaced using the mean-replacement method in order to subject all 34 items of the CAS to exploratory factor analysis. There were six eigenvalues > 1.0, but only three components suggested by the Scree Test (Cattell, 1966). Thus, we reran the analyses requesting the extraction of three components, using oblique Direct Oblimin rotation to maximize the clarity of the solution while allowing loadings onto secondary factors. This solution is shown in Table 1, together with internal consistencies (Cronbach a’s), and descriptive statistics. In line with Maltby et al. (2006), the components were labeled Entertainment-social, Intense-personal, and Borderline-pathological, and the average item score was computed for each component. The correlation between Entertainment-social and Intense-personal was r = .74; the correlation between Entertainment-social and Borderline-pathological was r = .51, and; the correlation between Borderline-pathological and Intense-personal was r = .69.

330

Structural Equation Modeling

331 332 333 334 335 336 337 338

Next, SEM was used to assess sex and age differences in the CAS factors. Given that the three components of the CAS were highly intercorrelated, we also modeled a latent, overall CAS factor. The two exogenous variables were age and sex (which were not intercorrelated) and they each had paths to self-rated attractiveness (the mediator), which, in turn, had a path to the overall factor of CAS. The model’s goodness of fit was assessed via the v2 statistic, the goodness-of-fit index

(GFI) and its adjusted version (AGFI), as well as the root mean-square residual (RMSEA) and the parsimony goodness-of-fit index (PGFI) (Kelloway, 1998; Loehlin, 1987; Maruyama, 1998). The hypothesized model did not fit the data well: v2 (df = 7, N = 781) = 51.2, p < .01, GFI = 0.97, AGFI = 0.94, RMSEA = 0.09 (low = 0.06, high = 0.11), PGFI = 0.32. The paths from sex to general CAS (0.00) and selfrated attractiveness (0.07) were not significant and were, therefore, deleted. On the basis of the modification indices, we added a path from sex to Entertainment-social. This path represents the effects of sex on Entertainment-social while controlling for overall CAS. The modified model (shown in Figure 1) fitted the data well: v2 (df = 8, N = 781) = 37.8, p < .01, GFI = 0.99, AGFI = 0.98, RMSEA = 0.06 (low = 0.04, high = 0.09), PGFI = 0.40, and accounted for 10% of the variance in CAS. Although the v2 value was significant, in large samples this is not unusual even in well-fitting models (Joreskog & Sorbom, 1993). Model trimming was used to test whether self-rated attractiveness mediated the path from age to CAS. Thus, the direct path from age to CAS was removed in order to assess the comparative fit of that model, which did not fit the data well: v2 (df = 9, N = 781) = 89.1, p < .01, GFI = 0.96, AGFI = 0.91, RMSEA = 0.11 (low = 0.08, high = 0.12), PGFI = 0.41.

339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364

Multigroup: Testing for Invariance Across Ethnicity

365 366

Finally, multigroup analysis was carried out in order to test whether the model was invariant across ethnic groups, that is between Malay and Chinese participants. The baseline model as shown in Figure 1 did not fit the Malay data well: v2 (df = 8, N = 512) = 44.1, p < .01, GFI = 0.97, AGFI = 0.93, RMSEA = 0.09 (low = 0.06, high = 0.12), PGFI = 0.38. Based on the modification indices, a path was added between self-rated attractiveness and Entertainment-social, which caused the model to fit the data well: v2 (df = 7, N = 512) = 22.8, p < .05, GFI = 0.99, AGFI = 0.96, RMSEA = 0.05 (low = 0.04, high = 0.09), PGFI = 0.38. The same model was tested on the Chinese sample and was found to fit the data well: v2 (df = 7, N = 269) = 4.7, p > .05, GFI = 0.99, AGFI = 0.98, RMSEA = 0.01 (low = 0.00, high = 0.05), PGFI = 0.37. Thus, the modified model (with a path from self-rated attractiveness to Entertainment-social) was used as the baseline. The unconstrained model fitted the data well: v2 (df = 14, N = 512 and 269) = 27.5, p < .05, GFI = 0.99, AGFI = 0.97, RMSEA = 0.03 (low = 0.01, high = 0.05), PGFI = 0.36, and paths were constrained accordingly. The fit indices for the constrained model were v2 (df = 21, N = 512 and 269) = 44.2, p < .01, GFI = 0.98, AGFI = 0.96, RMSEA = 0.03 (low = 0.01, high = 0.05), PGFI = 0.49, and the difference between the constrained and unconstrained model was significant: v2 (df = 7) = 16.74, p < .05. Based on the modification indices, the path between self-rated attractiveness and CAS was freed (unconstrained), after which the

367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394


286 287 288 289 290 291 292 293 294 295 296 297 298 299 300




5

Table 1. Rotated component matrix for the principal axis factor analysis with descriptive statistics and factor loadings

Items 16 18 15 17 19 20 13 21 14 33 3 7 6 12 8 4 1 5 34 2 27 26 29 32 23 30 22 25 10 28 11 24 9 31

Intense-personal

Borderline-pathological

10.12 20 .92 2.74 (0.91)

7.56 15 .89 2.19 (0.69)

10.4 14 .87 1.63 (0.55)

.90 .89 .86 .78 .74 .74 .66 .59 .42 .31


Rotated square loadings % variance explained Cronbach’s a Mean factor score and SD in brackets

Entertainment-social

.45

.40

.78 .75 .69 .57 .53 .52 .50 .48 .42 .40 .36 .41

.33

.31

.40 .75 .75 .71 .63 .59 .57 .56 .55 .51 .46

Note. N = 781. Extraction method: principal axis; rotation method: Direct Oblimin. Loadings < .30 not shown for the sake of clarity. Values in bold highlight secondary loading of the item. Actual items are available in Maltby et al. (2006).

395 396 397 398

structure of the model was invariant across both ethnic groups: difference in v2 (df = 6) = 7.61, p > .05. The final invariant model with coefficients for both groups is graphically depicted in Figure 2.

399

Further Comparisons

400 401 402 403 404 405

To examine ethnic differences in the CAS factor scores, we computed a series of Bonferroni-corrected (0.05/4 = 0.01) independent samples t tests with factor scores as the dependent variables and ethnicity as the independent variable. Results showed no significant ethnic difference on Entertainmentsocial (Malay M = 2.75, SD = 0.92; Chinese M = 2.73, Ó 2009 Hogrefe Publishing

SD = 0.91), t(779) = 0.20, p = .839, d = 0.02; Intensepersonal (Malay M = 2.21, SD = 0.70; Chinese M = 2.15, SD = 0.67), t(779) = 1.06, p = .290, d = 0.09; Borderlinepathological (Malay M = 1.59, SD = 0.53; Chinese M = 1.69, SD = 0.60), t(779) = 2.45, p = .020, d = 0.18; or on overall CAS (Malay M = 2.18, SD = 0.63; Chinese M = 2.19, SD = 0.64), t(779) = 0.23, p = .818, d = 0.02. Finally, to test for significant differences between factor scores, we computed a repeated-measures analysis of variance (ANOVA) with all participants. Results of Mauchly’s test of sphericity were significant, v2(2) = 240.42, p < .001, and for this reason the Greenhouse-Geisser correction was applied when interpreting the degrees of freedom. Results of the ANOVA showed a significant difference in European Psychologist 2010; Vol. xx(x):xxx–xxx

406 407 408 409 410 411 412 413 414 415 416 417 418 419

6



Figure 1. Modified model of age, sex, and self-rated attractiveness as predictors of celebrity worship. Note. *p < .01. Dotted path deleted from the hypothesized model; dashed path added to the hypothesized model. All coefficients are standardized beta values. Sex coded 1 = men, 2 = women. G CAS = General Celebrity Attitudes Scale component; SRA = self-rated attractiveness.

selected a politician. As Yue and colleagues (e.g., Cheung & Yue, 1999; Yue & Cheung, 2000a) have argued, this pattern of celebrity worship may reflect the fact that selection is driven by consumerism, superficiality, and illusory romance in some cultural contexts. More specifically, celebrity worship in Malaysia appears to be driven by an admiration for the commercial and hedonic success of celebrities, shaped (in part at least) by market and media forces (Yue & Cheung, 2000a). This can be distinguished from ‘‘model worship’’ cultures such as that found among mainland Chinese participants, where the moral and personality attributes of celebrities are emphasized and promoted. In this sense, our data support the idea that social differences in nationhood may affect celebrity selection in particular and the expression of celebrity worship in general. In cultures or nations where consumerism, illusory success, and hedonism are driving forces, it may be expected that individuals will show a preference for idols or celebrities rather than role models that demonstrate value sets absent in ‘‘three stars worship’’ cultures. In terms of the CAS, the results of our data reduction showed that the 34 items of the CAS could be reduced to three components, which we named Entertainment-social, Intense-personal, and Borderline-pathological. As such, the factor structure that emerged from the present data was consistent with previous studies conducted in the United States and Britain (e.g., Maltby et al., 2002, 2003, 2004, 2006), further supporting our conclusion that patterns of celebrity worship in Malaysia mirror that found in the West and in Hong Kong. These results would also seem to support Maltby et al.’s (2006, p. 282) contention that, in using the CAS, researchers will ‘‘always be able to identify sets of items that suggest entertainment-social, intense-personal, and borderline-pathological reasons for celebrity worship.’’ Of course, in the present study, we did not collect data from a Western setting, making direct comparisons difficulty. Even so, our results suggest that, despite cross-national differences in the symbolic meaning of celebrities across cultures (McCracken, 1989), celebrity worship itself appears to consist of three components that become more extreme, at least in ‘‘three stars worship’’ cultural settings. Importantly, our data showed that there were no significant differences between Malay and Chinese participants in mean ratings of the three CAS factors or overall CAS scores. Combined with the invariance of the model depicted in Figure 2 across ethnic groups, our results support previous work showing that the CAS has a three-dimensional factor structure (e.g., Maltby et al., 2002, 2003, 2004, 2006). Of course, caution should be applied when comparing the results derived from non-Western university students with samples recruited in the West. Specifically, our operationalization of ‘‘culture’’ in the present study was relatively freewheeling, and possibly obscures other relevant sociocultural factors that may influence celebrity worship in different national or cultural settings (cf. Hilton & Liu, 2008; Valsiner & Rosa, 2007). Nevertheless, in the present study, as in previous work (e.g., Maltby et al., 2004, 2006; McCutcheon et al., 2002), the three components of the CAS were highly correlated with one another. This supported the calculation of an overall CAS

Figure 2. Modified model invariant across ethnic groups. Note. *p < .01. Dashed path noninvariant across Malay and Chinese samples. All coefficients are standardized beta values. Coefficients for the Malay sample in italics, coefficients for the Chinese sample in bold. Sex coded 1 = men, 2 = women. G CAS = General Celebrity Attitudes Scale component; SRA = self-rated attractiveness. 420 421 422 423 424 425 426 427

the mean factor scores, F(1.58, 1232.67) = 1149.04, p < .001, gp2 = .60. Tests of simple effects showed that mean Entertainment-social ratings were significantly higher than both Intense-personal, t(780) = 24.91, p < .001, d = 0.68, and Borderline-pathological, t(780) = 39.35, p < .001, d = 1.47. In addition, Intense-personal scores were significantly higher than Borderline-pathological ratings, t(780) = 31.06, p < .001, d = 0.89.

428

Discussion

429 430 431 432 433 434 435 436 437 438 439

The aims of the present study were twofold: first, to examine patterns of celebrity selection among university students in Malaysian, and second, to examine the factor structure of the CAS in a non-Western sample. In the first instance, our results suggest that Malaysia may indeed be a ‘‘three stars worship’’ culture (Yue & Cheung, 2000a, p. 97), to the extent that both Malay and Chinese participants (despite differences in religious affiliation) overwhelmingly selected pop stars and movie stars as their favorite celebrities. Across ethnic groups, more than 75% of participants selected movie or pop stars as their favorite celebrity, while less than 5% European Psychologist 2010; Vol. xx(x):xxx–xxx


440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499


score, which we were then able to use in SEM. Results of this analysis for the total sample showed, firstly, that age was negatively associated with overall celebrity worship. This is consistent with previous work showing that celebrity worship becomes weaker with age (e.g., Ashe & McCutcheon, 2001; Giles, 2002), possibly because the aspects of celebrity worship associated with identity development are no longer as important after adolescence (Yue & Cheung, 2000a). Interestingly, however, the present results suggest that women were more likely to have higher scores on Entertainment-social than men. Previous studies have generally reported no sex biases in CAS components, although McCutcheon et al. (2004) did report that men had higher scores on Borderline-pathological than women. Although our results seem incongruent with previous work, it should be remembered that the association between participant sex and Entertainment-social in the present study was weak. Even so, we would recommend that future work examine the associations between sex and celebrity worship more systematically, and in particular develop a priori theorizing that can drive hypothesis testing. The results of our analyses also suggest that self-rated attractiveness is positively associated with celebrity worship. That is, individuals who rated themselves higher in physical attractiveness were more likely to have higher celebrity worship scores, possibly explained by higher scores on the Entertainment-social component. One possible explanation for this result rests on reported associations between self-rated attractiveness, personality, and celebrity worship. Specifically, Swami et al. (2008) have reported a significant positive association between self-rated attractiveness and the personality dimension of Extraversion, whereas Maltby et al. (2003) reported a positive association between Extraversion and Entertainment-social. It is possible, therefore, that the association between self-rated attractiveness and celebrity worship is mediated through personality attributes, and future work could examine this possibility in more detail. Finally, it is important to note that the model depicted in Figure 1 was not invariant across ethnic groups. Differences between Malay and Chinese participants, however, were relatively small (see Figure 2) and did not affect the overall factor structure of the CAS (i.e., the three-factor structure of the CAS was invariant across ethnic groups). Overall, therefore, it would appear that ethnicity only modestly moderates the associations between age, self-rated attractiveness, sex, and celebrity worship. As a final cautionary note, however, it should be highlighted that our model explained only a small proportion of the variance in CAS, suggesting that there may be other, more reliable factors that predict celebrity worship. For example, future work could examine whether the relationship between celebrity worship and such factors as exposure to different types of media and participants’ beliefs and attitudinal systems (e.g., consumerism, perceived illusory romance or success, and hedonism). Other limitations of the present study relate to the nature of the sample. First, we relied on a sample (albeit a large sample) of undergraduate students, and our results may not be generalizable either to the general population or to adolescent groups, for whom celebrity worship may be more pronounced. Second, we have only sampled Malay and

Chinese participants, but future work would do well to include other ethnic groups that can be found in Malaysia, or with first-generation migrants. Similarly and pace the discussion of cultural effects above, future research may also find it useful to replicate the present study with crosscultural or cross-national samples. A related problem is that the presentation of the CAS in Malay may have affected results between ethnic groups. Specifically, Malay may not have been the mother tongue for Chinese participants (as it would have been for most, if not all, Malay participants), although we stress that this is unlikely to have been a major limitation as Malay is the national language in Malaysia and is widely spoken among all ethnic groups (particularly at university level). Finally, the effect of age on celebrity worship may have been weakened by the restricted age range of participants in this study, again requiring replication of this study with the general population. In conclusion, the present preliminary results suggest that Malaysia may be a ‘‘three stars worship’’ culture, where market and media forces help determine the content of celebrity worship. In addition, our results suggest that the three-factor structure of the CAS may be relatively stable across different cultural and ethnic milieus, variations between ‘‘three stars worship’’ and ‘‘model worship’’ cultures notwithstanding. As such, our results support previous work using the CAS and, more generally, support McCutcheon et al.’s (2002) proposed ‘‘absorption-addiction’’ model of celebrity worship. Research in this area has benefited greatly from the development of the CAS, and our results further suggest that the instrument is a useful tool for investigating celebrity worship among Malay-speaking populations.

560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591

References

592

Abdul Wahab, J. (2006). Communication technology and the television industry in Malaysia. Paper presented at the ARC Asia Pacific Future Network International Conference, Perth, November 27–29. Agrawal, J., & Kamakura, W. A. (1995). The economic worth of celebrity endorsers: An event study analysis. Journal of Marketing, 59, 56–63. Alberoni, F. (1972). The powerless ‘‘elite’’: Theory and sociological research on the phenomenon of the stars. In D. McQuail (Ed.), Sociology of mass communications: Selected readings (pp. 75–98). Middlesex: Penguin. Ashe, D. D., & McCutcheon, L. E. (2001). Shyness, loneliness, and attitude towards celebrities. Current Research in Social Psychology, 6, 124–133. Boorstin, D. (1961). The image, or what happened to the American dream? New York, NY: Atheneum. Bernstein, I. H., & Teng, G. (1989). Factoring items and factoring scales are different: Spurious evidence for multidimensionality due to item categorization. Psychological Bulletin, 105, 467–477. Breslin, R. W. (1970). Back-translation for cross-cultural research. Cross-Cultural Psychology, 1, 185–216. Cattell, R. B. (1966). The scree test for the number of factors. Multivariate Behavioral Research, 1, 245–276. Cheung, C.-K., & Yue, X. D. (2000). Idol worshipping for vain glory, illusory romance or intellectual learning: A study in Nanjing and Hong Kong. International Journal of Youth and Adolescence, 8, 299–317.

593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620


500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559

7



8

Cheung, C.-K., & Yue, X. D. (2003). Identity achievement and idol worship among teenagers in Hong Kong. International Journal of Youth and Adolescence, 11, 1–26. Dyer, R. (1979). Stars. London: BFI. Erdogan, B. Z. (1999). Celebrity endorsement: A literature review. Journal of Marketing Management, 15, 291–314. Frommer, J., Juetteman-Lembke, A., Stratkoetter, A., & Tress, W. (1995). Subjective theories of illness and personality in depressive neurosis. Nervenarzt, 66, 521–531. Giles, D. C. (2000). Illusions of immortality: A psychology of fame and celebrity. London: MacMillan. Giles, D. C. (2002). Parasocial interaction: A review of the literature and a model for future research. Media Psychology, 4, 279–302. Giles, D. C., & Maltby, J. (2004). The role of media in adolescent development: Relations between autonomy, attachment, and interest in celebrities. Personality and Individual Differences, 36, 813–822. Henderson-King, D., Henderson-King, E., & Hoffman, L. (2001). Media images and women’s self-evaluations: Social context and importance of attractiveness as moderators. Personality and Social Psychology Bulletin, 27, 1407–1416. Hewstone, M., & Ward, C. (1985). Ethnocentrism and causal attribution in southeast Asia. Journal of Personality and Social Psychology, 48, 614–623. Hilton, D. J., & Liu, J. H. (2008). Culture and intergroup relations: The role of social representations of history. In R. Sorrentino & S. Yamaguchi (Eds.), Handbook of motivation and cognition across cultures (pp. 343–368). San Diego, CA: Academic Press. Houran, J., Navik, S., & Zerrusen, K. (2005). Boundary functioning in celebrity worshippers. Personality and Individual Differences, 38, 237–248. Johansson, S. (2005). Editorial. Westminster Papers in Communication and Culture, 2, 1–5. Joreskog, K. G., & Sorbom, D. (1993). LISREL 8: Structural equation modelling with the SIMPLIS command language. Chicago, IL: Scientific Software International. Kelloway, E. K. (1998). Using LISREL for structural equation modeling: A researcher’s guide. Thousand Oaks, CA: Sage. Loehlin, J. C. (1987). Latent variable models: An introduction to factor, path, and structural analysis. Hillsdale, NJ: Erlbaum. Maltby, J., Day, L., McCutcheon, L. E., Gillett, R., Houran, J., & Ashe, D. D. (2004). Personality and coping: A context for examining celebrity worship and mental health. British Journal of Psychology, 95, 411–428. Maltby, J., Day, L., McCutcheon, L. E., Houran, J., & Ashe, D. (2006). Extreme celebrity worship, fantasy proneness and dissociation: Developing the measurement and understanding of celebrity worship within a clinical personality context. Personality and Individual Differences, 40, 273–283. Maltby, J., Giles, D. C., Barber, L., & McCutcheon, L. E. (2005). Intense-personal celebrity worship and body image: Evidence of a link among female adolescents. British Journal of Health Psychology, 10, 17–32. Maltby, J., Houran, J., Lange, R., Ashe, D., & McCutcheon, L. E. (2002). Thou shalt worship not other gods – Unless they are celebrities: The relationship between celebrity worship and religious orientation. Personality and Individual Differences, 32, 1157–1172. Maltby, J., Houran, J., & McCutcheon, L. E. (2003). Locating celebrity worship within Eysenck’s personality dimensions. Journal of Nervous and Mental Disease, 191, 25–29. Marsden, P. (1997, November). Crash contagion and the death of Diana: Memetics as a new paradigm for understanding mass behaviour. University of Sussex, Brighton, Sussex, UK: Paper presented at the conference ‘‘Death and Diana’’. Maruyama, G. (1998). Basics of structural equation modeling. Thousand Oaks, CA: Sage.

Mathur, L. K., Mather, I., & Rangan, N. (1997). The wealth effects associated with a celebrity endorser: The Michael Jordan phenomenon. Journal of Advertising Research, 37, 67–73. McCracken, G. (1989). Who is the celebrity endorser? Cultural foundations of the endorsement process. Journal of Consumer Research, 16, 310–321. McCutcheon, L. E., Lange, R., & Houran, J. (2002). Conceptualization and measurement of celebrity worship. British Journal of Psychology, 93, 67–87. North, A. C., Sheridan, L., Maltby, J., & Gillett, R. (2007). Attributional style, self-esteem, and celebrity worship. Media Psychology, 9, 291–308. Petrides, K. V., Jackson, C. J., Furnham, A., & Levine, S. Z. (2004). Development of a short form of the Eysenck personality profiler via structural equation modelling. In K. van Monford, H. Oud, & A. Satorra (Eds.), Recent developments in structural equation modeling: Theory and applications. Dordrecht, NL: Kluwer Academic. Praet, C. (2002). Japanese advertising: The world’s number one celebrity showcase? A cross-cultural comparison of the frequency of celebrity appearances in TV Advertising. In M. S. Roberts & R. L. King (Eds.), International conference proceedings of the American Academy of Advertising (pp. 32–42). Gainesville, FL: American Academy of Advertising. Raviv, A., Bar-Tal, D., Raviv, A., & Ben-Horin, A. (1996). Adolescent idolization of pop singers: Causes, expressions, and reliance. Journal of Youth and Adolescence, 25, 631–650. Rojek, C. (2001). Celebrity. London: Reaktion Books. Schaller, M. (1997). The psychological consequences of fame: Three tests of the self-consciousness hypothesis. Journal of Personality, 65, 291–309. Swami, V. (2008). Translation and validation of the Malay Subjective Happiness Scale. Social Indicators Research, 88, 347–353. Swami, V. (in press). Psychometric analysis of the Malay version of the UCLA Loneliness Scale (ULS-8) and a comparison of loneliness among sojourning and non-sojourning Malaysian students. International Journal of Culture and Mental Health. Swami, V., & Chamorro-Premuzic, T. (2008). Factor structure of the Body Appreciation Scale among Malaysian women. Body Image, 5, 409–413. Swami, V., & Chamorro-Premuzic, T. (2009). Psychometric evaluation of the Malay Satisfaction with Life Scale. Social Indicators Research, 92, 25–33. Swami, V., & Furnham, A. (2008). The psychology of physical attraction. London: Routledge. Swami, V., Furnham, A., Georgiades, C., & Pang, L. (2007). Evaluating self and partner physical attractiveness. Body Image, 4, 97–101. Swami, V., Sinniah, D., Subramaniam, P., Pillai, S. K., Kannan, K., & Chamorro-Premuzic, T. (2008). An exploration of the Indecisiveness Scale in multi-ethnic Malaysia. Journal of Cross-Cultural Psychology, 39, 309–316. Swami, V., Taylor, R., & Carvalho, C. (in press). Acceptance of cosmetic surgery and celebrity worship: Evidence of associations among female undergraduate. Personality and Individual Differences. Swami, V., Taylor, R., & Carvalho, C. (2009). Associations between body dissatisfaction, sociocultural attitudes toward appearance, celebrity worship, the Big Five personality factors, and weight status. Manuscript in preparation. Swami, V., & Toveé, M. J. (2005a). Female physical attractiveness in Britain and Malaysia: A cross-cultural study. Body Image, 2, 115–128. Swami, V., & Toveé, M. J. (2005b). Male physical attractiveness in Britain and Malaysia: A cross-cultural study. Body Image, 2, 383–393.


621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689




690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758


789 790 791 792 793 794

Till, B. D., & Shimp, T. A. (1998). Endorsers in advertising: The case of negative celebrity information. Journal of Advertising, 27, 67–82. Turner, G. (2004). Understanding celebrity. London: Sage. Turner, G., Bonner, F., & Marshall, P. (2000). Fame games: The production of celebrity in Australia. Cambridge: Cambridge University Press. Valsiner, J. & Rosa, A. (Eds.). (2007). The Cambridge handbook of sociocultural psychology. Cambridge: Cambridge: University Press. Vigano, C. (1996). A case of TV erotomania. Journal of European Psychoanalysis, 2, 39–49. Yue, X. D. (1999). Comparing idol worship and model emulation in youth and adolescence. Journal of Youth Studies, 4, 137–152. Yue, X. D. (2000). Selection and admiration of four kinds of idols and models among young people: A comparative study among high school students in Hong Kong, Nanjing, Changsha, and Nantong. Journal of Youth Studies, 3, 153–168. Yue, X. D., & Cheung, C.-K. (2000a). Selection of favourite idols and models among Chinese young people: A comparative study in Hong Kong and Nanjing. International Journal of Behavioral Development, 24, 91–98. Yue, X. D., & Cheung, C.-K. (2000b). Idol worshipping and model learning among Chinese young people: A comparative study among students in Hong Kong, Guangzhou, and Changsha. Journal of Youth Studies, 3, 133–145.

Accepted September 25, 2009 About the authors

Subash K. Pillai is currently a Senior Lecturer at the University of Malaya Medical Center. His research interests are focused on issues related to child and adolescent psychiatry, mainly schizophrenia among adolescents.

795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817

Dr. Viren Swami

818

Department of Psychology University of Westminster 309 Regent Street London W1B 2UW UK Tel. +44 20 7911 5000 ext. 2120 E-mail [email protected]

819 820 821 822 823 824 825 826

Tomas Chamorro-Premuzic is a Senior Lecturer in Psychology at Goldsmiths, University of London. His main research area is differential psychology or individual differences, with particular interests in personality traits, cognitive abilities, leadership, and creativity. Khairul Mastor is based at the Centre for General Studies, University Kebangsaan Malaysia, and he has worked on various projects, including a study on lexical analyses. Jas Jaafar is the current head of the Department of Anthropology and Sociology, University of Malaya. Her research interests are wide ranging, including adolescent and child psychology, and general well-being. Dhachayani Sinniah is a master’s candidate in clinical psychology at the National University of Malaysia. Her research interests include suicidology and mental health among adolescents.


759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788

9

Viren Swami is a Senior Lecturer in Psychology at the University of Westminster. His main research interests are on interpersonal attraction from a cross-cultural perspective and predictors of positive body image.

