(PS3-68) Development of the Japanese Version of the Multidimensional Test Anxiety Scale (MTAS-J): Examination of Its Reliability and Validity

Abstract Text: [Introduction]
Test anxiety negatively impacts academic performance and mental health (e.g., Putwain et al., 2021). Despite numerous questionnaires measuring test anxiety have been developed, often with the same construct expressed in different terms. Therefore, it is necessary to develop a scale that can measure test anxiety from multiple components with reliability and validity. This study aimed to develop the Japanese version of the Multidimensional Test Anxiety Scale (MTAS-J; Putwain et al., 2021) and examine its psychometric properties.

[Method]
A survey was conducted with Japanese high school students via their homeroom teachers, resulting in 1,332 valid responses (615 males, 687 females, 30 others). Participants completed: (1) MTAS-J, (2) established test anxiety measures: TAI (Araki, 1989; Spielberger et al., 1980), TAS (Sakano, 1988; Sarason, 1972), (3) anxiety: Short-SCAS (Ishikawa et al., 2018), (4) depression: DSRS-C (Birleson, 1981; Murata et al., 1996), (5) school adjustment: SLAQ (Ladd, Kochenderfer & Coleman, 1997; Otai et al., 2014), and (6) self-reported end-of-term examination scores (English and mathematics). The study recieved approval from Doshisha University's Ethics Committee (Approval No. 202401).

[Results]
Confirmatory factor analysis supported both the four-factor structure model (worry, cognitive interference, tension, and physiological arousal) and the higher-order factor model, with acceptable fit indices (CFIs > .90, TLIs > .90, RMSEAs < .08, SRMRs < .05). Internal consistency was satisfactory across all measures (αs > .70, ωs > .70, λ6s > .70). Test-retest reliability was satisfactory for the total score (r = .70, ICC = .71) but not for the individual subscales (rs < .70, ICCs < .70). Regarding validity, both simple correlations and latent two-variable correlations showed that the MTAS-J exhibited patterns similar to the original MTAS. Strong correlations were observed between established measures (TAI/TAS) and the MTAS-J total score (TAI: r = .82, ρ = .86; TAS: r = .76, ρ = .88) and subscale scores (rs > .50, ρs > .50). The MTAS-J showed strong correlations with anxiety, moderate correlations with depression, and weak to no correlations with school adjustment. Final examination scores showed weak negative correlations with the total score of the MTAS-J scores and moderate negative correlations with the cognitive interference subscale.

[Discussion]
The MTAS-J demonstrated acceptable internal consistency, factor validity, and adequate test-retest reliability for the total score. These findings suggest that the MTAS-J is a valid instrument for assessing the multifaceted nature of test anxiety among Japanese students. Furthermore, the significant associations between the MTAS-J and measures of anxiety, depression, school adjustment, and test performance provide evidence of criterion-construct validity. The MTAS-J will serve as a valuable tool for both basic research and intervention studies addressing adolescent mental health and academic performance.