A Multi-Method Multi-Analytic Approach to Establishing Internal Construct Validity Evidence: The Sport Multidimensional Perfectionism Scale 2

This article presents a chronology of three empirical studies that outline the measurement process by which two new subscales (Doubts about Actions and Organization) were developed and integrated into a revised version of Dunn, Causgrove Dunn, and Syrotuik's (2002) Sport Multidimensional Perfectionism Scale (i. e., the Sport-MPS-2). All studies were designed to provide internal validity evidence that supported the inclusion of these new subscales into the instrument. Phase 1 determined that expert judges regarded the new Doubts about Actions and Organization items (n = 12) as possessing adequate levels of content-relevance and content-representativeness. Phases 2 and 3 examined the latent dimensionality of the Sport-MPS-2 using multidimensional scaling and factor analytic techniques. At a structural level, results indicated that the new Doubts about Actions and Organization subscales were suitable for inclusion in the Sport-MPS-2. Initial external validity evidence is presented via correlational and regression analyses surrounding the relationships between Sport-MPS-2 subscales and global self-esteem. The benefits of utilizing a multi-method multi-analytic approach in the initial stages of the construct validation process are discussed.