Student Academic Involvement: A Comparison of Classical Test Theory and Item Response Theory

This study compares the application of classical test theory versus item response theory in the validation of students' academic engagement questionnaire. The sample size of this research was 750 students (265 boys and 485 girls) who were selected by multi-stage cluster sampling method among the students of Tehran universities. Due to the non-availability of the Persian version of the questionnaire, first its translation and content analysis was done by experts. Then, its validation was done through exploratory and confirmatory factor analysis and its reliability using ordinal theta. In the following, clean power and acceptability, as well as the curve of the question characteristics and the awareness function of the whole test were also checked through R software. The analysis of the question parameters and the subjects determined that this questionnaire is able to distinguish between people in terms of the concept of conflict. Also, based on classical test theory and question theory, it is possible to determine which questions better measure social involvement and academic involvement. According to the classical test theory, the accuracy of the two questionnaires based on the ordinal theta is suitable and the classical test theory cannot show and compare how people in higher and lower groups answer the questions of social involvement and academic involvement. Finally, based on the sum of the fit indices, it can be seen that the data are fit with the assumed model.