Examination of measurement invariance of PISA-2018 Bullying Exposure Scale across countries: Türkiye, Mexico, Finland sample

The aim of this research is to examine the measurement invariance of the bullying exposure scale located at the bottom of the PISA 2018 cycle student survey across different country groups. The sample of the research consists of individuals responding to the scale from Türkiye, Mexico and Finland. Initially, the analysis started with 15151 data, and after testing the assumptions, extreme values were removed from the data set and the study continued with 13005 data. The bullying exposure scale used as a data collection tool consists of 6 items. The bullying exposure model was created using Exploratory Factor Analysis. In the resulting model, 6 items were grouped under 2 factors and the model was confirmed by Confirmatory Factor Analysis. Afterwards, measurement invariance stages across different countries were examined sequentially with Multi-Group Confirmatory Factor Analysis. For this process, it was checked whether the fit indices were within the acceptable range at each stage. Additionally, changes in CFI and TLI values were examined. Intercountry comparisons were made between Türkiye-Mexico, Türkiye-Finland and Mexico-Finland groups. In the Türkiye-Mexico and Turkey-Finland models, the configural invariance and metric invariance stages were achieved, but the scalar invariance stage was not achieved. In the Mexico-Finland model, the stages of configural invariance, metric invariance and scalar invariance are achieved. As a result, measurement invariance of the bullying exposure scale across different country groups could not be achieved.