PERTANIKA JOURNAL OF TROPICAL AGRICULTURAL SCIENCE

A psychometrically sound measurement instrument is crucialy in understanding needed athletes' psychological profile. Among indices of a psychometrically sound instrument are its factor structure and internal consistency. The present study examined the factorial validity and reliability of the Malaysian version of the Revised Competitive State Anxiety Inventory - 2 using confirmatory factor analysis (CFA) and Cronbach alpha, respectively. The questionnaire was administered to 267 Taekwondo athletes. The sample consisted of 58% boys and 42% girls. Mean age for the overall sample was 13.2±3.01. Four models were tested: a 1-factor, a 2-factor, a 3-factor models and a 3-factor model with correlated error terms. The 3-factor model was predicted to produce the best model fit consisting of somatic anxiety, self-confidence, and cognitive anxiety components. Confirmatory Factor Analysis results revealed a theoretically meaningful and close model fit of the 3-factor model (χ² = 223.13, df = 116, df/ χ² = 1.92, CFI = 0.92, TLI = 0.91, RMSEA = 0.05) compared to the other three models. The results also showed an acceptable level of alpha coefficient for the subscales (α = 0.78 for somatic anxiety, 0.76 for cognitive anxiety and 0.83 for self-confidence subscales). However, a slightly high relationship was observed between somatic and cognitive anxiety subscales (r = 0.81), indicating an issue with the subscales discriminant validity. In conclusion, the results generally support the factorial structure and internal consistency of the Malaysian version of CSAI-2R but future analysis is still needed to confirm the findings.