Résumé
The purpose of the present study was to test the validity of the Child Sport Cohesion Questionnaire (CSCQ) via convergent, discriminant, predictive, and factorial validity. Children (N = 290, Mage = 10.73 ± 1.13 years) from six elementary schools completed questionnaires assessing cohesion, satisfaction, anxiety (cognitive and somatic), and self-efficacy, as well as items concerning sport type (interactive and independent) and team tenure (length of time on team). Task and social cohesion demonstrated significant (p < .05) positive correlations to satisfaction and significant (p < .05) negative correlations to anxiety (cognitive and somatic), providing support for convergent validity. Discriminant validity was demonstrated in that task cohesion had a significantly stronger (p < .01) relationship with self-efficacy than did social cohesion. As for predictive validity, no significant differences (p >.05) were found between sport type (interactive and independent) and cohesion (task and social). Social cohesion was however significantly greater (p <.05) in longer standing compared to more recent team members (i.e., team tenure). Therefore, only partial support for predictive validity was found. Finally, factorial validity was demonstrated with a strong model fit (CFI = .96, TLI = .95, RMSEA = .07, SRMR = .04), a moderate inter-factor correlation (r = .53), and good internal consistency values for task (?= .90) and social (?= .90) cohesion.Acknowledgments: Research funded by the Social Sciences and Humanities Research Council of Canada