Refining the Canadian Assessment of Physical Literacy based on theory and factor analyses

Background: The Canadian Assessment of Physical Literacy (CAPL) is a 25-indicator assessment tool comprising four domains of physical literacy: (1) Physical Competence, (2) Daily Behaviour, (3) Motivation and Confidence, and (4) Knowledge and Understanding. The purpose of this study was to re-examine the factor structure of CAPL scores and the relative weight of each domain for an overall physical literacy factor. Our goal was to maximize content representation, and reduce construct irrelevant variance and participant burden, to inform the development of CAPL-2 (a revised, shorter, and theoretically stronger version of CAPL). Methods: Canadian children (n = 10,034; M-age = 10.6, SD = 1.2; 50.1% girls) completed CAPL testing at one time point. Confirmatory factor analysis was used. Results: Based on weak factor loadings (lambda s < 0.32) and conceptual alignment, we removed body mass index, waist circumference, sit-and-reach flexibility, and grip strength as indicators of Physical Competence. Based on the factor loading (lambda < 0.35) and conceptual alignment, we removed screen time as an indicator of Daily Behaviour. To reduce redundancy, we removed children's activity compared to other children as an indicator of Motivation and Confidence. Based on low factor loadings (lambda s < 0.35) and conceptual alignment, we removed knowledge of screen time guidelines, what it means to be healthy, how to improve fitness, activity preferences, and physical activity safety gear indicators from the Knowledge and Understanding domain. The final refined CAPL model was comprised of 14 indicators, and the four-factor correlated model fit the data well (r ranged from 0.08 to 0.76), albeit with an unexpected cross-loading from Daily Behaviour to knowledge of physical activity guidelines (mean- and variance-adjusted weighted least square [WLSMV]chi(2)((70)) = 1221.29, p < 0.001, Comparative Fit Index [CFI] = 0.947, root mean square error of approximation [RMSEA] = 0.041[0.039, 0.043]). Finally, our higher-order model with Physical Literacy as a factor with indicators of Physical Competence (lambda = 0.68), Daily Behaviour (lambda = 0.91), Motivation and Confidence (lambda = 0.80), and Knowledge and Understanding (lambda = 0.21) fit the data well. Conclusions: The scores from the revised and much shorter 14-indicator model of CAPL can be used to assess the four correlated domains of physical literacy and/or a higher-order aggregate physical literacy factor. The results of this investigation will inform the development of CAPL-2.