Educational Big Data Analytic – A Mediation Analysis of the Covariates of Academic Performance

Abstract

Family issues have been acknowledged a common challenge faced by students, with a significant impact on academic performance among the students. However, the world statistics revealed a decline in student’s academic performance due to the increasing number of family issues and these causes are possibly linked to educational and individual factors in each student, affecting student’s academic performance. This study investigates the influence of family-related factors on students' academic performance, focusing on parental socioeconomic status, mental health, distance from home to school, and environmental factors such as the place they lived. The educational big datasets were collected from ICPSR’s National Longitudinal Study of Adolescent to Adult Health (>90,000 respondents, 42 datasets) in analysing family issues and its impacts on education, and statistical analyses were used to identify and examine the significant associations between these factors and academic outcomes. Three most relevant datasets between SPSS are used to analyse the correlation between academic performances with various factors. IBM SPSS 23.0 and macro PROCESS 4.2, and statistical analyses were used to identify significant associations between these factors and academic outcomes. The findings suggest that students of higher socioeconomic backgrounds, with better mental health, living closer to the school and in more developed countries tend to have better academic performance. Besides, negative correlation was found between mental health and academic performance. These findings are believed to provide important insights for the complex relationship between family-related factors and academic performance, contributing to better understanding and concerns about factors that affect students’ academic success for educators and researchers.