In obesity modelling studies, researchers have been seeking to identify the effective indicators of obesity by using appropriate statistical or mathematical techniques. The main objective of the present study is addressed in three stages. First, a new framework for modelling obesity in university students is introduced. The second stage involves data analysis based on Bayesian Structural Equation Modelling (BSEM) for estimating the Body Mass Index (BMI) (representative of the obesity level) of students at three university levels: Bachelor, Master and PhD. In the third stage, the highest significant correlation is determined between the BMI and other variables in the research model that were found significant through the second phase. The data for this study were collected from students at selected Malaysian universities. The results indicate that unhealthy food intake (fast food and soft drinks), social media use and stress exhibit the highest weightage contributing to overweight and obesity issues for Malaysian university students.

The emergency transition from physical to online learning during COVID-19 has affected university students in various aspects, especially their academic performance. It can be caused by many factors, such as individual, environmental and social factors. Therefore, this study aims to determine the impact of fear, stress, well-being, teacher and parents’ support (independent variables) on undergraduates’ academic performance (dependent variable) during the COVID-19 pandemic. A structured online questionnaire has been developed and administered to 400 undergraduates. A structural equation model that integrated all variables under investigation was built and statistically validated using AMOS. The results demonstrated that well-being, teacher emotional support and teacher academic support have the highest significant impact on the respondents’ academic performance. It can be concluded that teachers’ support is the most substantial influence in ensuring student learning sustainability during the COVID-19 pandemic.

Bayesian Structural Equation Modeling (SEM-Bayesian) was applied across different research areas to model the correlation between manifest and latent variables. The primary purpose of this study is to introduce a new framework of complexity to adolescent obesity modeling based on adolescent lifestyle through the application of SEM-Bayesian. The introduced model was designed based on the relationships among several factors: household socioeconomic status, healthy food intake, unhealthy food intake, lifestyle, body mass index (BMI) and body fat. One of the main contributions of this study is from considering both BMI and body fat as dependent variables. To demonstrate the reliability of the model, especially in terms of its fitting and accuracy, real-time data were extracted and analyzed across 881 adolescents from secondary schools in Tehran, Iran. The output of this study may be helpful for researchers who are interested in adolescent obesity modeling based on the lifestyle and household socioeconomic status of adolescents.

No abstract

No abstract