Network Analysis for Educational Data Mining and Visualization: Interrelationships of Problem Behavior Symptoms Among Rural Children

Adolescent problem behavior is closely related to their academic achievement as well as their physical and psychological development. The effects of children's problem behaviors have been examined in prior studies, typically as separate variables. However, limited attention has been given to the interconnections among the symptoms of these behaviors. We address this gap in existing literature by utilizing both directed and undirected network analysis, construct the network structure and find out the potential direction of association between problem behavior symptoms. The study focused on 1777 rural children (Mage = 11.26 years, SD = 1.29). The results showed that the three most central symptoms of problem behaviors are “Smoking”, “Stealing at home”, and “Secret drinking”. “Smoking” is considered the trigger symptom, and “Unrepentance” is considered the final symptom in the network. The results of this study offer insights for preventing and intervening problem behavior among adolescent in rural area. The study uses data processing and visualization methods. It helps explain complex problems in adolescent education.