A penta-dimensional longitudinal analysis of the predictors of compulsive internet use among adolescents using linear mixed model (LMM)

In the present study, we investigated the relationships between demographics, socio-psychological factors, substance abuse, sexual behavior, and types of internet use, collectively called the “penta-dimension”, and compulsive internet use (CIU) among adolescents. Foremost, the research was inspired by the vast proliferation of internet usage through desktop computers and their internet-accessible mobile derivatives (laptops, cellular phones, and tablet computers) among adolescents, plus their ever-increasing amount of time spend in the virtual world, which have been shown to have detrimental effects on their livelihoods. Second, the research was an attempt to not only centralize the ever-so fragmented research fields of internet addiction, but also to explore the unknown. ❧ We obtained data from the sixth funded project (TND-6), the seventh clustered-randomized controlled trial of Project Towards No Drug Abuse (TND). The results show four of the five dimensions were associated with CIU: demographics, socio-psychological factors, sexual behavior, and types of internet use. Substance abuse failed to predict CIU. Specifically, our study revealed eight significant predictors of compulsive internet use behavior among adolescents: gender, ethnicity, BMI, parental education level (a measure of SES), behavior problems, decision making–avoidance, sex frequency in past 12 months, and online communication. In order to construct a model of adolescent CIU, more studies specifically designed for CIU are needed. During our research process, however, we may have accidentally stridden into a new dimension, rarely discussed within the realm of adolescent CIU research.