Psychometric Properties and Measurement Invariance of a Torrance-Inspired Test of Creative Thinking in Primary School Pupils

Creative thinking is increasingly recognised as an essential educational competency, yet psychometrically robust measurement tools validated for specific cultural contexts remain scarce. This study evaluates the structural validity, internal consistency, discriminant validity, and measurement invariance of a Torrance-inspired Test of Creative Thinking (TCT) among primary school pupils (N = 302, ages 5–12). Confirmatory factor analyses demonstrated adequate fit for a four-factor structure (Fluency, Flexibility, Elaboration, Originality) for the figural component of the battery. However, discriminant validity analyses revealed high inter-factor correlations among fluency, flexibility, and elaboration, suggesting they operate as closely coordinated facets of divergent thinking, whereas originality demonstrated psychometric independence. Multigroup analyses established scalar invariance across biological sex, indicating equivalent measurement functioning between male and female pupils. Conversely, metric invariance failed across age cohorts. Multiple-indicator multiple-cause (MIMIC) modelling identified significant uniform differential item functioning (DIF) across several indicators, suggesting that developmental differences influence task interaction. The findings provide psychometric support for assessing creativity in primary education but caution against direct score comparisons across diverse age groups without accounting for developmental measurement variance.