Norms and Factorial Invariance of the Athletic Identity Measurement Scale (AIMS)

Abstract

This study examined the factorial invariance of the Athletic Identity Measurement Scale (AIMS; Brewer, Van Raalte, & Linder, 1993) and developed norms for its future use. A sample of 2,856 participants was assembled from multiple administrations of the AIMS over the past 10 years. The total sample was divided into two samples: a derivation sample (n = 1,462) and a validation sample (n = 1,394). The samples were stratified to insure equivalence in gender and varsity athlete/non-athlete composition. No significant differences were detected between the derivation and validation samples in AIMS item mean scores or demographic variables (all ps>.10). Results indicated that a multidimensional model in which three highly correlated first order factors (social identity, exclusivity, and negative affectivity) are subordinate to a higher order athletic identity factor demonstrated factorial invariance across genders and athletic statuses. Confirmatory factor analytic (CFA) methods were performed on the derivation sample data to examine four AIMS models (the original uni-dimensional model and three multidimensional models, designated Models A, B, C, and D) that have been previously proposed and investigated in the literature. The results of these analyses, which are described in the Appendix, supported the development of a new higher-order model, Model E, which consists of seven items. The findings suggested that a 7-item composite AIMS score is appropriate for assessing athletic identity in both men and women. Applications of the AIMS with student-athletes were discussed.