Measuring TPACK in context: Validating a cross-contextually fair instrument for Indonesian teachers

Abstrak

As technology integration has become increasingly central to contemporary education, the need for valid and contextually fair instruments for assessing teachers' Technological Pedagogical Content Knowledge (TPACK) has become more urgent. This study aimed to develop and comprehensively validate a TPACK self-assessment instrument for Indonesian in-service teachers. A total of 1,297 teachers selected through purposive and convenience sampling from diverse provinces, school levels, and geographic locations participated in the study. The 48-item instrument, distributed across seven TPACK dimensions (TK, CK, PK, PCK, TCK, TPK, and TPACK), employed a 4-point Likert scale. Psychometric evaluation followed a multistage approach: (a) construct validity via corrected item-total correlations, in which all items met the r ≥ .30 threshold (range: .57–.79); (b) internal consistency reliability using Cronbach's alpha, with all dimensions exceeding .70 (α = .78–.97); (c) exploratory factor analysis (EFA), which supported the theoretical factor structure (KMO = .968; Bartlett's χ² = 54,279.51, p < .001); (d) confirmatory factor analysis (CFA), which yielded marginal fit indices (CFI = .707; RMSEA = .107), suggesting the need for model refinement; and (e) differential item functioning (DIF) analysis, which revealed minimal gender-based bias but notable location-based DIF. Overall, the instrument demonstrated strong psychometric properties for measuring teachers' TPACK in the Indonesian context. The findings highlight the importance of refining TPACK instruments to ensure greater measurement fairness across diverse educational settings and support future teacher professional development initiatives.