AUTHOR=Silva-Rocha Viviane Vedovato , de Sousa Diogo Araújo , Osório Flávia L. TITLE=Psychometric Properties of the Brazilian Version of the Sport Anxiety Scale-2 JOURNAL=Frontiers in Psychology VOLUME=10 YEAR=2019 URL=https://www.frontiersin.org/journals/psychology/articles/10.3389/fpsyg.2019.00806 DOI=10.3389/fpsyg.2019.00806 ISSN=1664-1078 ABSTRACT=

Competitive anxiety (CA) is an emotional reaction manifested at a somatic and/or cognitive level that regularly appears before or during sports competitions and can significantly impact an athlete’s performance. Given the scarcity of validated instruments available for evaluating the competitive-anxiety trait in the Brazilian context, this study aimed to investigate the psychometric properties of the Sport Anxiety Scale-2 (SAS-2). The study sample was composed of 238 professional and amateur athletes aged 13 years or older who practice different sports modalities. The results of confirmatory factor analysis (CFA) revealed adequate fit indices of the original three-factor theoretical model of the SAS-2 after including a correlation between the errors for items 6 and 12 of the somatic anxiety subscale (CFI = 0.97, TLI = 0.96, RMSEA = 0.08, WRMR = 1.04). For convergent and divergent validity, the SAS-2 subscales exhibited a positive and strong correlations with the Competitive State Anxiety Inventory-2R (CSAI-2R; r = 0.52–0.82), weak to moderate correlations with the State-Trait Anxiety Inventory – only the trait scale (STAI-T; r = 0.49–0.59), weak correlations with the Social Phobia Inventory (SPIN; r = 0.29–0.41) and weak to moderate correlations with the Patient Health Questionnaire (PHQ-9; r = 0.49–0.56). The SAS-2 was also able to discriminate among participants with and without social anxiety, general trait anxiety and depressive symptoms, thus confirming its discriminant validity. According to ROC curve analysis, the cutoff point at a score of 29 indicated the optimal balance of sensitivity (0.74) and specificity (0.82). The internal consistency (α = 0.73–0.86) and the test-retest reliability (ICC = 0.73–0.80) were satisfactory. These results indicated that the Brazilian version of the SAS-2 exhibited satisfactory psychometric performance and could be used in the Brazilian context.