Development of a domain-specific measure of perfectionism in sport.

摘要

The purposes of this dissertation are to (a) develop sport-based versions of Frost, Marten, Lahart, and Rosenblate's (1990) doubts about actions and organization subscales, and (b) produce construct validity evidence to support the inclusion of these new subscales in a revised version of Dunn, Causgrove Dunn, and Syrotuik's (2002) Sport multidimensional Perfectionism Scale (Sport-MPS). Efforts to achieve these purposes are presented in a series of phases based upon Messick's (1989) conceptualization of validity. The first phase (Chapters 2 and 3) presents the development of sport-based doubts about actions (DAA-Sport) and organization (ORG-Sport) domain specifications. The second phase (Chapter 4) describes the construction of DAA-Sport and ORG-Sport items and establishes content-related validity evidence for the new items via an analysis of expert judges' ratings of item content relevance and item-set content representativeness. The third phase (Chapter 5) presents structurally-related validity evidence for the DAA-Sport and ORG-Sport subscales through multidimensional scaling (MDS) analyses of 33 elite-level Ultimate Frisbee players' similarity ratings between DAA-Sport items, ORG-Sport items, and Sport-MPS items. Results indicate the unique nature of DAA-Sport and ORG-Sport within the context of original Sport-MPS items. These first three phases indicate that DAA-Sport and ORG-Sport are suitable to be included in a revised version of the Sport-MPS (i.e., the Sport-MPS-2). The fourth phase (Chapter 6) presents an examination of the latent dimensionality of the Sport-MPS-2 through a factor analytic examination of 251 Canadian Intercollegiate Sport student-athletes' responses to the instrument. Results indicate that the Sport-MPS-2 is best represented by six factors (i.e., Personal Standards, Concern Over Mistakes, Perceived Parental Pressure, Perceived Coach Pressure, Doubts About Actions, and Organization). The final phase (Chapter 7) establishes external validity evidence for the Sport-MPS-2 through an examination of the relationships between 181 male intercollegiate varsity ice hockey players' Sport-MPS-2 scores and their scores on global perfectionism and competitive trait anxiety measures. Analyses indicate that Sport-MPS-2 subscales are related in theoretically meaningful ways to both measures. The construct validity evidence present in this dissertation provides initial support for including DAA-Sport and ORG-Sport subscales into the Sport-MPS-2 for the purpose of examining perfectionism in sport.