The investigation of measurement invariance of TIMSS 2019 student bullying scales according to gender and culture

The purpose of this research is to investigate the measurement invariance of Trends in Mathematics and Science Study (TIMSS) 2019 grade 4 and grade 8 student bullying scales according to gender and culture (Russia, Chinese Taipei, Türkiye). The sample of the research consists of 4th and 8th grade students participating in the TIMSS 2019 research from Russia, Chinese Taipei and Türkiye. Within the scope of the research, public use version of TIMSS data was used. The data collection tools of this research are TIMSS 2019 grade 4 and grade 8 student bullying scales. First, Principle Components Analysis (PCA) results and Cronbach Alpha internal consistency coefficients as proof of reliability were shared for each country. Then, the single-factor structure of the scales was confirmed by Confirmatory Factor Analysis (CFA) for each country. The Measurement Invariance study was carried out by calculating Multiple Group Confirmatory Factor Analysis (MGCFA), one of the Structural Equation Modeling (SEM) techniques. While the strict invariance model of the TIMSS 2019 grade 4 student bullying scale was justified for Russia, Chinese Taipei and Türkiye according to gender, only the configural invariance model of the Grade 8 student bullying scale according to gender was justified. In the comparisons made according to culture, only the configural invariance model was provided for both grade levels. Consequently, TIMSS 2019 grade 8 student bullying scale does not provide measurement invariance for Russia, Chinese Taipei and Turkey according to gender and culture. TIMSS 2019 grade 4 student bullying scale provided measurement invariance for Russia, Chinese Taipei and Turkey according to gender, but it did not provide it for those countries according to the culture. Key Words: TIMSS, Measurement Invariance, Student Bullying Scale