Validation of the Persian version of the Internet Addiction Questionnaire: Comparison of Multiple Indicator Multiple Causal Model (MIMIC) with Multiple Group Multiple Indicator Multiple Causal Model (MG-MIMIC)

Objective(s) : It is believed that frequent use of Interne might be harmful for health. The Internet Addiction Test (IAT) is one of the most practical tools for diagnosing Internet addiction. The aim of the present study was to investigate validation of the Persian version of the questionnaire using Differential Item Functioning (DIF) analysis. DIF evaluates how people in different subgroups comprehend the same questionnaire item. Methods : A cross section study on a sample of 317 students of Mashhad University of Medical was conducted. To collect data, the Internet Addiction Questionnaire and demographic information and Behavioral Features Checklist were administered. Multiple Indicator Multiple Causal Model (MIMIC) and Multiple Group Multiple Indicator Multiple Causal Model (MG-MIMIC) were used to investigate the differentiating of items. Results: The mean score of the IAT questionnaire was 40.92±11.70, which based on the cut-off point of 46, 29.5% of students were addicted to the Internet. The results of the MIMIC models indicated that there was no difference in each of the IAT questionnaire items. Also, considering the grouping structure did not have much effect on identifying DIF, because the results of the MG-MIMIC model were almost the same as the MIMIC model. Conclusion : Due to the lack of significant DIF in IAT questions and having measurement invariance this study can be considered as empirical support for the IAT questionnaire as a valid and structurally stable tool.