The youth market has been characterized as one of the most coveted market segments because of its spending power and tremendous potential for becoming lifetime customers (Bush, Martin, & Bush, 2004). The product market for adolescents is expected to grow to more than $208 billion by 2011, according to a report from market research firm Packaged Facts (Sass, 2007). Apparel, jewelry, and cosmetics are top product categories for adolescent girls and are important products used by adolescents to portray personal identity (Ossorio, 1995). Because of adolescents' buying power and the important role apparel plays in adolescents' lives, it is important for apparel marketers to understand the adolescent consumer market.

The purpose of this research was to examine the relationship between each of three independent variables (i.e., reference groups, clothing symbolism, and clothing involvement) and adolescents' clothing purchase intentions and clothing behavior, and whether the three variables are predictors of adolescents' clothing purchase intentions and clothing behavior. Reference groups are a source of instrumental and emotional support, offering adolescents a sense of belonging during their physical, emotional, and cognitive adjustment (Blackwell, Miniard, & Engel, 2001). Adolescents may have many different types of reference groups. The reference groups included in the currents study were friends, popular girls, and parents. Specifically in the current study, ninth grade girls' perceptions of friends' clothing behavior, popular girls' clothing behavior, and parents' opinions concerning clothing behavior were examined for their relationship to the ninth grade girls' clothing purchase intentions and behavior. The second independent variable was clothing symbolism. Adolescents may use clothing as a symbol (i.e., clothing symbolism) to express their actual self-concepts or to attain their ideal self-concepts (Erickson, 1983; Solomon & Rabolt, 2004). Two types of clothing symbolism were included in the study: the degree of congruity between actual self-concept and the perceived images of four outfits (i.e., actual self and clothing image congruity) and the degree of congruity between ideal self-concept and the perceived images of four outfits (i.e., ideal self and clothing image congruity). The third independent variable was clothing involvement. Viera (2009) found that young consumers are highly involved with clothing. The degree of clothing involvement may be closely related to adolescent girls' clothing purchase intentions and their clothing behavior.

A conceptual model that formed the framework for this study was developed by integrating several theories, propositions, and research findings in the literature. Based on the framework, 16 research questions were formulated. Focus groups provided input for questionnaire development, and four outfit images, one each considered sexy, conservative, springy, or sporty, were identified and included in the questionnaire. Before the main data collection, the questionnaire was pilot tested and revised. Data collection was conducted at three high schools in central Virginia, and 353 female students in the ninth grade participated. Standard and stepwise multiple regression analyses were used to address the research questions.

Among the four outfit images, that with a sexy image was found to have the highest mean score for ideal self-concept, indicating that participants would most like to view themselves sexy. Results of a factor analysis for clothing involvement revealed three factors: clothing importance, clothing expressions, and clothing brand perceptions. Participants had fairly high mean scores for all three clothing involvement factors. Among those three factors, clothing importance had the highest mean. In addition, results showed that participants perceived that their friends and popular girls would most often wear the outfit with a sporty image, and they also perceived that their parents would most like them to wear an outfit with a sporty image. The sporty outfit image also had the highest mean for participants' clothing purchase intentions and clothing behavior among the four outfit images.

Results of one of the stepwise regressions, for the sexy outfit image showed that perceptions of reference groups' clothing behavior or opinions (i.e., friends' clothing behavior, parents' opinions concerning clothing behavior, popular girls' clothing behavior), one type of clothing symbolism (i.e., ideal self and clothing image congruity), and one clothing involvement factor (i.e., clothing importance) predicted adolescent girls' purchase intentions for the outfit with a sexy image. The results for wearing clothing with a sexy image when hanging out with friends were slightly different from those for purchase intentions. For the behavior of wearing an outfit with a sexy image, the variable "perceptions of popular girls' clothing behavior" was not a predictor, but actual self and clothing image congruity was. For the springy outfit image, all the perceptions of reference groups' clothing behavior or opinions and actual self and clothing image congruity were the best predictors of participants' clothing behavior; however only friends' clothing behavior and popular girls' clothing behavior were significant predictors of adolescent girls' purchase intentions for this outfit image. For the conservative and sporty outfit images, only the perceptions of reference groups' clothing behavior or opinions predicted adolescent girls' clothing purchase intentions and clothing behavior.

In conclusion, the research findings suggest a powerful relationship between adolescent girls' perceptions of reference groups' clothing behavior or opinions about clothing behavior and the girls' own clothing behavior and purchase intentions. Participants' perceptions of reference groups' clothing behavior or opinions were the best predictors of the participants' clothing behavior and purchase intentions for all four outfits. Furthermore, clothing symbolism (e.g., using a sexy outfit to express or attain a sexy image) appears to motivate ninth grade girls to wear clothing with a sexy or springy image. The girls would wear a sexy outfit to portray their actual self-concepts and attain their ideal self-concepts. They also would wear outfits with a springy image to portray themselves. Additionally, the more the participants in this study considered clothing to be important, the more likely they were to purchase and wear a sexy image outfit; however the participants indicated that, of the four outfit images in the study, they most purchase and wear clothing like the sporty image outfit the most for hanging out with friends in comparison to the other three outfit images. Based on the findings, suggestions and implications for parents, educators, and marketers were provided.