This study examined the valid items in the perception of TPACK competencies of physics, math, biology, and chemistry teacher candidates in one public university in Indonesia. This study used a cross-sectional survey where data were collected once from one sample at a time. The study was conducted in four teacher education programs in one public university. This study used a stratified proportioned random sampling technique. The sample for this study consisted of 400 students from 4 teacher education programs including Physics Education, Chemistry Education, Biology Education, and Mathematics programs. Data were collected through a questionnaire with 75 statements related to TPACK competency indicators. The data were processed and analyzed through Partial Least Squares Structural Equation Modeling (PLS-SEM) and by using SmartPLS 3.0 software. The results showed that of 75 statements, there were 27 valid statements related to TPACK competency indicators. These 27 statements can potentially be used to measure TPACK competencies of teacher candidates. The results obtained from the factor analysis and the Smart PLS procedures showed that the construction of items in the instrument and their relevance to each other is appropriate. Based on the overall results, one of the main findings of this study is that the predictive power of the core knowledge base and the second-level knowledge base on TPACK development is significantly different.
Read full abstract