Validation of the Amharic version of Internet Addiction Test-20: a cross-sectional study

Abstract

Background: Internet Addiction is defined as excessive internet use or poorly controlled preoccupations, impulses, or behaviors related to computer use and internet access that cause impairment or suffering. It had devastating effect on people lives, families, productivity, academic performance and rarely engaging in criminal acts like alcohol use, drug addiction, or compulsive gambling. This study aimed to investigate the psychometric properties of the Amharic version of Internet Addiction Test-20 among Addis Ababa University, College of Health Sciences medical students, Addis Ababa, Ethiopia. Methods: A cross sectional study was carried out among 410 medical students using a convenience sampling method after stratifying them based on their year level. SPSS Version 23 was used to do Pearson’s correlation coefficient to determine the convergent validity of Amharic version of IAT. We computed correlation coefficient between the aggregate scores of IAT-20 and the scores for depressive symptoms, problematic substance use, and other characteristics of participants which was assessed using Patient health questionnaire-9 (PHQ-9), Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) and questionnaire developed to assess demographic and internet use related characteristics, respectively. AMOS 23 software was used to conduct confirmatory factor analysis (CFA) to evaluate the construct validity of Amharic version of IAT. Test–retest reliability was also determined with 2 weeks interval (n = 51). Results: The data confirmed a two-factor structure. Normed Fit Index (NFI) = 0.89, Tucker Lewis Index (TLI) = 0.91 and Comparative Fit Index (CFI) = 0.92, Root Mean Square Error Approximation (RMSEA) = 0.07, and Standardized Root Mean Residual (SRMR) =0.05 indicated a good fit model structure. There was moderate positive correlation between the aggregate scores of IAT-20 and PHQ-9 scores (r = 0.55, p < 0.00), but weak positive correlation between IAT-20 and ASSIST scores (r = 0.14, p < 0.00). IAT-20 was also found to have good internal consistency (Cronbach’s alpha = 0.89 for each factor). The test–retest reliability was also good for all items (Intra Class Correlation Coefficient (ICC) > 0.30) except item 16. Conclusion: We found that the IAT-20 is psychometrically sound and a simple screening test for Internet Addiction. However, it is important to acknowledge that further studies are necessary to replicate these findings on diverse population.