Age and Gender Effects in Physical Self-Concepts: A Longitudinal Comparison between Athletes and Nonathletes during Preadolescence

Age and Gender Effects in Physical Self-Concepts: A Longitudinal Comparison between Athletes and Nonathletes during Preadolescence Marco Guicciardi, and Massimiliano Pastore Department of Psychology, University of Cagliari, Italy Previous researches outlined that self-concept becomes with age more differentiated and more highly correlated with external indicators of physical competence. Despite the bodily change, the physical self-concept appears to be relatively stable during the preadolescence. Following previous studies, physical self-concepts were predicted to be substantially higher for athletes than nonathletes and substantially higher for men than women: gender differences were predicted to be smaller for athletes than nonathletes (Marsh, 1998). A longitudinal study was carried out to evaluate the development of self-concept during preadolescence. All participants, males and females, were students enrolled in middle school classes in Sardinia, attending a three-year motor program, aimed to support the development of coordinative motor skills. Some students have practiced sport activities out of school time. In a multiple occasion design five waves of physical self concept responses were collected by means of the Physical Self Description Questionnaire (PSDQ), developed by Marsh et al. (1994) and previously translated and validated in Italian language by Meleddu, Scalas and Guicciardi (2002). Longitudinal analysis performed using Lisrel models confirms the stability of the physical self concept (NNFI = .91; CFI = .93; RMSEA = .104) and offers some evidences to support the success of motor program.

Introduction During the preadolescence girls and boys experiment very important modifications in their bodies. The increase of the bodily dimensions, the sexual maturation and the rapid rhythm of growth due to the puberty claim new psychomotor adjustments. Sport activity can contribute to sustain self confidence and to provide pieces for the construction of a stable identity. Sport allows self-affirmation through physical performance, security in dealing with the body and opportunity for the development of social and emotional relationship with peers (Brettschneider and Heim, 1997). These attainments are very important in this stage because often the uncertainty of movements couples with a consistent decline in many dimensions of the self-concept. As Marsh (1989) stated self concept declines during adolescence showing a quadratic effect that was consistent for both genders. Nevertheless, the physical self-concept appears, to be relatively stable during this potentially turbulent period (Marsh 1998). Hypotheses According to previous studies, physical self-concepts were predicted to be substantially higher for athletes than nonathletes and substantially higher for men than women (Marsh, 1996). With age the self concept becomes more differentiated and more highly correlated with external indicators of physical competence: we expected gender stereotypes affected preadolescent self-concepts with larger differences for facets connected with the sub-domain of the Physical Ability than Appearance (Marsh, 1989; Fox 1997). Because sport experience stabilizes changes in the self concept (Guicciardi and Polidetti, 2003), Gender differences were predicted to be smaller for athletes than nonathletes (Marsh, 1998; Brettschneider 1999) and structural relationships were predicted to be more stables over time at the higher levels (Shavelson, Hubner, Stanton, 1976). Methods A longitudinal study was carried out to evaluate the development of self-concept in the preadolescence. In a multiple occasion design, five waves of physical self concept responses were collected over a three-years period of time by means of the Physical Self Description Questionnaire (PSDQ), developed by Marsh et al. (1994) and previously translated and validated in Italian language by Meleddu, Scalas and Guicciardi (2002). The PSDQ was administered at the start of 1st classes (6th grades), at the start and the end of 2nd classes (7th grades) and at the start and the end of the 3rd classes. Only half portion of the initial group classes completed data in all occasions and was retained as sample for the present study. Participants were 58 (29 boys and 29 girls) middle school students observed from years 6 (mostly 11 years of age) to years 8 (mostly 13 years of age), from three different classes of two schools in Cagliari and Sinnai, Italy. All participants, boys and girls, attended a three-year motor program, realized by the PE teachers with the contribution of the CONI technical staff, aimed to support the development of coordination motor skills. Over the threeyears period, half of the sample practiced frequently or regularly sport activities other than compulsory physical education classes. Results In preliminary analyses, mean differences in PSDQ scales were evaluated. A profile analyses was performed using gender (males/females), group (athlete/nonathlete) and time (the five testing occasions) as design variables and PSDQ

scales as observed variables. This double multivariate analysis of variance shown a large multivariate effects for group (Wilks’ Lambda = .39, F(11,44) =6.27, p