Abstract
Self-regulated learning (SRL) refers to athletes' active engagement in their own practice via planning, monitoring, and adapting processes (Zimmerman, 1986). SRL has been positioned as an individual difference variable impacting expertise development (Tedesqui & Young, 2015). McCardle et al. (2017) validated the structural validity of athletes' SRL self-report survey – the SRL-SRS for Sport Training (Bartulovic et al., 2017). Their measurement model, which also included earlier items from Toering et al.'s (2012) survey, showed acceptable model fit and divergent validity. This investigation aimed to examine the predictive validity of the same SRL-SRS for Sport Training model using skill level as a criterion outcome. Canadian athletes (n = 369; age 13 - 58 years) completed 53 SRL items and reported their highest performance level (local/regional, provincial, national, international). Multivariate analyses of variance tested for skill level differences on the constituent SRL processes (i.e., planning, checking, evaluating-reflecting, effort, self-efficacy) assessed in the survey resulting in significant differences: Wilk's ? = .915, F (15, 947) = 2.07, p = .009, partial ?2 = .029. Follow up discriminant analyses showed differences between the international athletes; self-efficacy and effort were the strongest contributors to the discriminant function, with evaluating-reflecting also contributing. A pattern emerged where local/regional athletes reported more engagement in many SRL processes than provincial athletes, and more SRL on certain processes than national athletes. Results are discussed in terms of remaining steps in SRL-SRS validation, measurement development, and limitations that may constrain effect sizes.Acknowledgments: This work was supported by a Social Sciences and Humanities Research Council of Canada Insight Development Grant 430-2015-00904 (Bradley W. Young, PI).