The phenomenon of idolization is a characteristic of adolescence. An idol is someone whose talents, achievements, status, or physical appearance are especially recognized and appreciated by his or her fans. Since objects of idolization come from various domains of life, sport champions, movie actors/actresses, television personalities, pop stars, and political or religious leaders can all be idolized figures (Raviv, Bar-tal, Raviv, & Ben-Horn, 1996). Idolization can take many different forms. From extreme cases of suicides and skipping lessons to catch a glimpse of their idols to listening to CDs casually every now and then, adolescents show their devotion and admiration of their idols in various different ways (McCutcheon, Lange, & Houran, 2002).
Psychologists believe that during adolescents' transition to adulthood they form strong attachments to figures around them as they search for their own identity (Greene & Adams-Price, 1990; Raviv et al., 1996; Yue & Cheung, 2000). Under such circumstances, identification with some adults or peer idols enables adolescents to seek information or values and to prepare themselves for an adult role (Erikson, 1964; 1968). Despite this important phenomenon, research on idolization behavior is lacking (Cheng, 1997; Yue & Cheung, 2000).
There are several reasons why it may be important to study celebrity idolization (Engle & Kasser, 2005). First, the identity of adolescents is partially constructed by interacting with popular media and celebrities. As per Steele and Brown's (1995) conclusion that media activities are important determinants of adolescents' cognitions and behaviors and that media are sometimes even used to regulate mood, it is necessary to understand why and how idolization forms in adolescents. Second, idolization is an avenue by which they explore romantic views and attitudes toward interpersonal relationships, especially for girls. Karniol (2001) suggested that idolization may affect the types of relationships they form. Third, idolization may be involved in a subject's well-being. For instance, Cheng (1997) found that celebrity fan club members in Hong Kong reported lower self-esteem than did non-fan club members. Finally, idolization is clearly relevant to the economic behavior of adolescents.
Studies on idolization in Chinese societies are equally rare and have been mostly conducted in Hong Kong (So & Chan, 1992; Wong & Ma, 1997; Chan, Cheung, Lee, Leung, & Liu, 1998; Cheung & Yue, 1999; Yue & Cheung, 2000; see also Yue & Cheung, 2000 for a review). The purpose of this study is to clarify the social or personality attributes of an idol and a model as perceived by adolescents in a Chinese society--Taiwan. This exploratory study attempts to describe in depth the patterns of Taiwanese adolescents' idol worship behaviors. Specifically, we delineate the idol traits which attract adolescents, examine how adolescents rate their favorite idols based on underlying idol traits, and finally explore the relationship between idol traits and adolescents' worship levels.
A mail survey was conducted of students from 13 senior high schools across Taiwan; 2,100 questionnaires were mailed and 1,958 were returned. After deleting questionnaires with incomplete data, 1,636 were used for analysis. The response rate was 77.9%.
The questionnaire was designed to measure respondents' criteria for choosing favorite idols and models. It is composed of idol characteristics, reasons for worship, intensity of respondents' worship levels, and respondents' characteristics. Eight dichotomous items regarding reasons for worship were augmented from Chiang (2003). The worship levels were measured using the 17-item, 5-point Likert type celebrity worship scale (McCutcheon et al., 2002). The reliability of the celebrity worship scale estimated with Cronbach's alpha was 0.94. The worship levels (McCutcheon et al., 2002) suggest three increasingly extreme sets of idol worship behavior: individualistic behaviors, social activities, and borderline pathology. As worship level increases, these behaviors increasingly occur together. Since the celebrity worship construct is best considered as unidimensional, we calculate the mean score of the 17 items to represent respondents' intensity of worship levels.
RESULTS AND DISCUSSIONS
The sample of adolescents is composed of more females (83.3%) than males (16.7%), and the mean age is 17 years with a range from 15 to 19. Thirty-two percent of the adolescents are in their first level of senior high school, 32.8% in the second level, and 35.6% in the third level. The residential areas of the sample are 34.2% in northern Taiwan, 32.5% in central Taiwan, 23.9% in southern Taiwan, and 9.4% in eastern Taiwan. Overall, the sample appears to be adequate in representing Taiwanese senior high school adolescents' idol adoration behaviors. Table 1 summarizes the sample characteristics.
Adolescents' Favorite Idols
The gender of the adolescents' favorite idols is more male (65.4%) than female (34.6%). Consistent with Yue and Cheung's results in Hong Kong (2000), most (66.8%) of the selected favorite idols are celebrities in media or entertainment fields (e.g., actors, singers, or athletes). Some (10.1%) are noncelebrities (e.g., family, friends, or teachers), and some (10.7%) are fictitious characters (e.g., Spiderman), whereas few are politicians (3.0%), intellectuals (4.6%), or entrepreneurs (4.8%). According to the results, adolescents tend to worship idols who are accessible in their daily lives (e.g., noncelebrities) or through mass media (e.g., media stars and fictitious characters). As to the reasons for idolization, "Good-looking" was chosen most (41.4%), while "rich" was the least (9.0%). Table 2 summarizes the idol characteristics and reasons for idolization.
Does Gender Matter?
Since the data on idols' characteristics show a gender difference in terms of favorites, more extensive analyses were conducted. Results indicate that the gender of more than half of the adolescents (55.5%) chose is opposite from that of their favorite idols. Table 3 shows that gender of adolescents and idols are associated ([chi square] (1) = 29.16, p < 0.01). Female adolescents appear to prefer male idols (62.6%) to female idols (37.4%), whereas this phenomenon is the same for male adolescents, whose favorite idols also are more male (79.%) than female (20.4%). On the other hand, male adolescents adore male idols more (79.6%) than do female adolescents (62.6%).
We next analyzed the association between adolescents' gender and the types of favorite idols. As shown in Table 4, when the idol is female, adolescents' gender and idol types are not associated ([chi square] (5) = 1.32, p > 0.5). However, when the idol is male (see Table 5), adolescents' gender is significantly related to the idol types ([chi square] (5) = 27.54, p < 0.01). Table 5 shows that female adolescents worship more male media stars (69.2%) than do males (51.8%). By contrast, male adolescents worship more male politicians (7.3%) and more fictitious characters (19.3%) than do females (2.8% for politicians and 13.1% for fictitious characters).
Overall, the result is in line with that of Raviv et al. (1996)--that most of the male and female adolescents choose a male media star as their favorite rather than a female and rather than other types of idols. Despite the consistency between male and female adolescents' selections, the psychological processes are supposed to be distinct from each other. The findings for the female might be due to the fact that male media stars provide a safe and convenient romantic attachment for female adolescents (Adams-Price & Greene, 1990; Karniol, 2001), whereas identification attachment accounts for males' worship of male media stars (Adams-Price & Greene, 1990). Furthermore, when considering the absolute numbers in Tables 4 and 5, we found only 21 female fictitious idols (vs. 154 male fictitious idols) and only 9 female politicians (vs. 40 male politicians). The differences may reflect gender inequality in the virtual world and in the political field.
Underlying Idol Traits that Attract Adolescents
Because there are eight causes for worshipping idols, the next analysis revealed the major dimensions of variation between these reasons for worship. We defined the dichotomous items of reasons to worship as ordinal and thus applied categorical principal components analysis for quantifying the observed reasons. Table 6 summarizes the results. Four meaningful dimensions, accounting for 69% of the variance, were retained to represent four latent idol traits that attract adolescents.
On the first dimension, "good-looking," "attractive dressing," and "attractive body shape" have high and positive component loadings (all other worship causes have mediocre positive loadings). We refer to this dimension as "exterior." By contrast, the second dimension "interior," reflecting idols' inside virtues," correlates mainly with "humorous and funny," "knowledgeable and clever," "civic-minded and caring," and "good-looking," "attractive dressing," "attractive body shape," in opposite directions, while having little relation with "rich" and "sporty." This pattern suggests that idols with a high score on "interior" have a high score on "humorous and funny," "knowledgeable and clever," and "civic-minded and caring," and a low score on "good-looking," "attractive dressing," and "attractive body shape." "Wealth," the third dimension, reveals a contrast between "rich," "knowledgeable and clever," and "humorous and funny." Adolescents consider that wealthy idols are clever while also serious. The fourth dimension, "athlete," contrasts between "sporty" and "knowledgeable and clever."
The above analyses raise two issues. First, all reasons for idolization are positively loaded on the trait "exterior." This pattern implies that good-looking idols with attractive body shapes and wearing attractive apparel are rich, sporty, and also have good virtue. Could the "perfect idol" be merely an illusion? The "perfect idol syndrome" might be attributed to the halo effect of idols' brilliant images. Second, the "interior" trait contrasts between idols' good virtue and outside images, the "wealth" contrasts between rich and humorous, and the "athlete" contrasts between sporty and knowledgeable. Are those patterns realities or only stereotypes assigned to idols? Collectively, adolescents' evaluations of idols appear to be dominated by the halo emitted by idols' surface images. When these images are not prominent, adolescents seem to choose idols by stereotyping. To facilitate subsequent analyses, latent traits scores were calculated respectively for each respondent's favorite idol.
How are different Idol Types Evaluated on Idol Traits and Worship Levels?
We then conducted a MANOVA to examine how adolescents evaluate the six idol types on the four underlying idol traits as a whole. The results indicate that the six types show significantly different overall traits (Wilks's lambda = 0.66; Pillai's trace = 0.38; Hotelling's trace = 0.48; all ps < 0.01). We then conducted four independent univariate ANOVAs on these four traits, and the results show that different types of idols are considered differently on these traits, respectively (all ps < 0.01). Tukey's HSD post hoc comparisons were analyzed to determine the subtle differences. Table 7 summarizes the comparison results.
Media stars are regarded as having more favorable exteriors than any other types of idols. However, media stars exhibit a less pleasing interior and are less affluent than most other types of idols. Entrepreneurs, unsurprisingly, are the wealthiest across the six idol types, while noncelebrities have the highest interior trait. Interestingly, fictitious characters are thought of as being much wealthier and having a better interior than media stars.
ANOVAs on the worship levels show that the six types of idols are worshiped with different intensities (F(5, 1630) = 8.62, p < 0.01. Tukey's HSD post hoc tests (see Table 7) indicate that media stars are worshiped more intensely than intellectuals, entrepreneurs, and fictitious idols. We speculate that this phenomenon might be due to the intensive commercial manipulation to promote media idols (Giles, 2000).
Factors that Influence the Intensity of Worship Levels
A regression model was conducted to assess how well the set of idol traits explain adolescents' worship levels. As discussed, gender-dissimilarity-attraction and idol types are both critical variables that influence adolescents' worship behaviors; thus we also include these two variables in the regression model. Gender difference was coded as a dummy variable with one indicating a different gender between the adolescent and idol. The idol type is also coded as a dummy variable with one indicating that the type of the adolescent's favorite idol is a media star. Worship levels were then regressed on gender difference, idol type, exterior, interior, wealth, and athlete. Table 8 shows the regression results.
As can be seen in Table 8, except for the interior trait, all independent variables are significantly related to worship levels. We found that differences in sex between adolescents and their favorite idols was associated with adolescents' heightened worship levels. Adolescents are more likely to be fascinated with their idols when he/she is a media star or is attractive for the exterior or athlete traits. However, the worship levels are negatively correlated with the wealth trait, but are not associated with the interior trait.
We could derive further implications from the worship pattern exhibited by the model. First, though the four underlying idol traits appear to be positive in nature, their impacts on adolescents' worship levels are nonsymmetrical. While the traits of exterior and athlete are worship facilitators, the wealth trait is a prohibitor. Unexpectedly, the interior traits is neutral from adolescents' perspectives. Next, similar to the finding of Raviv et al. (1996), the exterior trait is the most important factor in adolescents' worship intensity. In combination with the illusory exterior (as discussed, the composition of exterior trait implies perfect idol illusion) and indeterminate interior, adolescents' worship behaviors appear to be nonrational. In other words, idolization may be related to materialism, which emphasizes the right image and possessions (Engle & Kasser, 2005), and adolescents may choose idealism (vs. realism) and romanticism (vs. rationalism) in selecting idols (Yue & Cheung, 2000).
To explore adolescents' idol adoration behavior, we collected 1,636 samples from 13 senior high schools across Taiwan. The results show: (a) over half of the adolescents adore gender idols that are different from themselves. The results also show that most female and male adolescents' favorite idols are male. Specifically, female adolescents tend to worship more male media stars than do male adolescents, while male adolescents worship more male politicians and more fictitious characters than do female adolescents. (b) To explore the reasons for idolization, the four idol traits that attract adolescents--exterior, interior, wealth, and athlete--were retained in this study. Interestingly, we found that those respondents tend to stereotype their idols. For example, adolescents thought that media stars would have a more favorable exterior trait, but would have less of an interior trait, entrepreneurs make up the wealthiest group, and noncelebrities have the highest interior trait. (c) The idolization level differs between idol types and provide different reasons for worship. The worship levels intensify when idols are media stars or when adolescents are attracted by an idol's exterior trait. However, the idolization levels are negatively correlated with the wealth trait and are not associated with the interior trait.
Adolescents tend to establish a positive self-identity and social identity through idolization (Cheng, 1997; Raviv et al., 1996). Self-categorization often occurs in the process of searching for an identity (Turner, 1987). Our findings imply that adolescents mainly classify themselves and their favorite idols using the exterior trait, which intensified idolization behaviors. Compared with the social value that often emphasizes an individual's interior trait, it seems that the trait of exterior is not a good basis for establishing adolescents' self-identity. Adolescents' selection of an idol is probably market-driven and media-determined (Yue & Cheung, 2000), and technology might be blamed for the spread of these delusional beliefs within the youth culture (Showalter, 1997). These behaviors based on idols' exterior traits may prevent adolescents from seeking information regarding other idols who are not exterior-oriented, thus increasing the stereotyping of those idols (i.e., those who are interior, wealthy, or athlete-oriented).
As individuals' idol worship attitudes and behaviors are negatively correlated with their cognitive abilities, such as critical thinking and creativity (McCutcheon, Ashe, Houran, & Maltby, 2003), one plausible explanation for our finding of distorted idol worship behavior is adolescents' cognitive deficits. Adolescents may be relatively low in cognitive functioning, which causes them to more easily accept commercial and materialistic messages (Engle & Kasser, 2005), thus precluding them from admiring an idol's interior. Adolescents' idol adoration behaviors also influence their consuming behaviors (Chiou, Huang, & Chuang, 2005) and sometimes have detrimental effects on their mental and physical health (Maltby, Giles, Barber, & McCutcheon, 2005). In addition, adolescents may select idols as their epistemic authorities on whom they rely in accumulating knowledge or forming values (Engle & Kasser, 2005; Raviv et al., 1996). Since this reliance is likely to be connected to adolescents' delinquent behaviors (Chan et al., 1998), educators should stress the importance of idolization behaviors, and make the public aware of mass media manipulation of adolescents' cognitive deficits.
Adams-Price, C., & Greene, A. L. (1990). Secondary attachment and adolescent self-concept. Sex Roles, 22(3/4), 187-198.
Chan, W. T., Cheung, C. K., Lee, T. Y., Leunj, K. K., & Liu, S. C. (1998). Moral values of youth in Hong Kong. Hong Kong: Department of Applied Social Studies, City University of Hong Kong.
Cheung, C. K., & Yue, X. D. (1999). Idol worshipping for vain glory, illusory romance or intellectual learning: A study in Nanjing and Hong Kong. International Journal of Adolescence and Youth, 7, 20-28.
Cheng, S. T. (1997). Psychological determinants of idolatry in adolescents. Adolescence, 32, 687-692.
Chiang, Y. M. (2003). A study of the relationship between idolization and self-concept. Unpublished Master's Thesis, National Taichung Teachers College, Taichung, Taiwan, R.O.C.
Chiou, J. S., Huang, C. Y., & Chuang, M. C. (2005). Antecedents of Taiwanese adolescents' purchase intention toward the merchandise of a celebrity: The moderating effect of celebrity adoration. The Journal of Social Psychology, 145(3), 317-332.
Engle, Y., & Kasser, T. (2005). Why do adolescent girls idolize male celebrities? Journal of Adolescent Research, 20(2), 263-283.
Erikson, E. (1964). Memorandum on identity and Negro youth. Journal of Social Issues, 20, 29-42.
Erikson, E. (1968). Identity: Youth and crisis. New York: W. W. Norton.
Giles, D. C. (2000). Illusions of immortality: A psychology of fame and celebrity. London: MacMillan.
Greene, A. L., & Adams-Price, C. (1990). Adolescents' secondary attachment to celebrity figures. Sex Roles, 23, 335-347.
Karniol, R. (2001). Adolescent females' idolization of male media stars as a transition into sexuality. Sex Roles, 44(1/2), 61-77.
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(1), 17-32.
McCutcheon, L. E., Ashe, D. D., Houran, J., & Maltby, J. (2003). A cognitive profile of individuals who tend to worship celebrities. The Journal of Psychology, 137(4), 309-322.
McCutcheon, L. E., Lange, R., & Houran, J. (2002). Conceptualization and measurement of celebrity worshiop. British Journal of Psychology, 93, 67-87.
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(5), 631-650.
Showalter, E. (1997). Hystories: Hysterical epidemics and modern media. New York: Columbia University Press.
So, C. Y. C., & Chan, J. M. (1992). Mass media and youth in Hong Kong: A study of media use, youth archetype, and media influence. Commission on Youth: Hong Kong Government.
Steele, J. R., & Brown, J. D. (1995). Adolescent room culture: Studying media in the context of everyday life. Journal of Youth and Adolescence, 24, 551-576.
Turner, J. C. (1987). Rediscovering the social group: A self-categorization theory. Oxford: Blackwell.
Wong, S. W., & Ma, K. (1997). A survey on the patterns of canto-pop appreciation in Hong Kong. Department of Applied Social Studies, City University of Hong Kong.
Yue, X. D., & Cheung, C. K. (2000). Selection of favorite idols and models among Chinese young people: A comparative study in Hong Kong and Nanjiing. International Journal of Behavioral Development, 24(1), 91-98.
The authors contributed equally to this research; the order of authorship was determined by a coin flip. The authors gratefully acknowledge Yen-Ju Chen in conducting the study and coding the data.
Chien-Hsin Lin, Department of International Business Studies, National Chi Nan University, Nantou, Taiwan, R.O.C. and Department of International Business, Yu Da College of Business, Miaoli, Taiwan, R.O.C.
Requests for reprints should be sent to Ying-Ching Lin, Department of Business Administration, National Dong Hwa University. 1, Sec. 2, Da Hsueh Rd., Shou-Feng #974, Hualien County, Taiwan, R.O.C. E-mail: firstname.lastname@example.org
Table 1: Sample Characteristics N % Gender Female 1,362 83.3 Male 274 16.7 Age (in Years) 15 28 1.7 16 411 25.1 17 635 38.8 18 506 30.9 19 56 3.4 Senior High School Level First 516 31.5 Second 537 32.8 Third 583 35.6 Residential Area Northern Taiwan 560 34.2 Central Taiwan 531 32.5 Southern Taiwan 391 23.9 Eastern Taiwan 154 9.4 Table 2: Idol Characteristics N % Gender Female 566 34.6 Male 1070 65.4 Idol Type Media Stars (e.g., actor, singer, or athlete) 1093 66.8 Politicians 49 3.0 Intellectual Idols (e.g., scientist, writer, 76 4.6 musician, or artist) Entrepreneur Idols 78 4.8 Non-Celebrities (e.g., family, friend, or 165 10.1 teacher) Fictitious Characters (e.g., Spiderman) 175 10.7 Reasons for Idolization Good-looking 677 41.4 Attractive dressing 370 22.6 Attractive body shape 368 22.5 Rich 148 9.0 Humorous and funny 529 32.3 Knowledgeable and clever 571 34.9 Sporty 365 22.3 Civic-minded and caring 385 23.5 Table 3: Adolescent's Gender vs. Favorite Idol's Gender Adolescent's Gender Idol's Gender Female Male Female 510 (37.4%) 56 (20.4%) Male 852 (62.6%) 218 (79.6%) [chi square] (1) = 29.16, p < 0.01 Table 4: Adolescent's Gender vs. Female Idol Type Adolescent's Gender Female Idol Type Female Male Media Stars 350 (68.6%) 40 (71.4%) Politicians 8 (1.6%) 1 (1.8%) Intellectual 24 (4.7%) 3 (5.4%) Entrepreneur 27 (5.3%) 2 (3.6%) Non-Celebrities 83 (16.3%) 7 (12.5%) Fictitious Characters 18 (3.5%) 3 (5.4%) [chi square] (5) = 1.32, p > 0.5 Table 5: Adolescent's Gender vs. Male Idol Type Adolescent's Gender Male Idol Type Female Male Media Stars 590 (69.2%) 113 (51.8%) Politicians 24 (2.8%) 16 (7.3%) Intellectual 37 (4.3%) 12 (5.5%) Entrepreneur 34 (4.0%) 15 (6.9%) Non-Celebrities 55 (6.5%) 20 (9.2%) Fictitious Characters 112 (13.1%) 42 (19.3%) [chi square] (5) = 27.54, p < 0.01 Table 6: Component Loadings of Worship Reasons on Idol Traits Idol Trait Worship Reason Exterior Interior Wealth Athlete Good-looking 0.654 -0.375 -0.239 -0.088 Attractive dressing 0.736 -0.275 -0.090 -0.180 Attractive body shape 0.727 -0.319 -0.022 -0.118 Humorous and funny 0.433 0.546 -0.304 0.179 Knowledgeable and 0.302 0.611 0.306 -0.399 clever Civic-minded and 0.473 0.541 -0.194 -0.044 caring Rich 0.393 -0.054 0.828 0.040 Sporty 0.389 0.062 0.115 0.833 Eigenvalue 2.31 1.29 0.99 0.94 of Variance Explained 28.80 16.1 12.3 11.8 Table 7: Results of Multiple Comparisons of Idol Traits and Worship Levels between Idol Types Idol Type Media Stars Politicians Intellectual Entrepreneur (A) (B) (C) (D) Dependent Variable Means (a) Variables Exterior 0.14 -0.53 -0.62 -0.30 Interior -0.21 0.41 0.31 0.03 Wealth -0.18 0.49 0.22 1.74 Athlete 0.06 -0.38 -0.43 -0.52 Worship Level 2.98 2.79 2.56 2.52 Idol Type Non-Celebrities Fictitious (E) (F) Dependent Variables Variable Means (a) Pair Comparisons Exterior -0.09 -0.26 A>B ***, A>C ** A>D ***, A>E **, A>F ***, E>C *** Interior 0.84 0.26 AF *** Wealth -0.04 0.17 AE ***, C
E ***, D>F *** Athlete 0.01 0.11 A>B **, A>C ***, A>D ***, B C ***, A>D ***, A>F **, Level 2.86 2.77 D …