A prediction model for peer attachment in KOREAN female adolescents using back propagation neural network

Abstract

Background/Objectives: This study used data mining technique to explore the potential factors affecting the peer attachment of South Korean female students.Methods/Statistical analysis: This study analyzed 2009 9th grade female students, who attended Panel Study on Korean Children in 2016. Peer attachment was defined as a dependent variable. The explanatory variables included gender, academic achievement satisfaction, subjective household economy level, parent-child dialogue frequency, subjective health status, depression symptom, self-esteem, subjective life satisfaction, and mobile phone dependence. The predictors of peer attachment were analyzed by using back propagation neural network (BPN).Findings: Analysis results showed that depression, self-esteem, dialogue level between parent and child regarding school life, subjective health condition were highly related to the peer attachment of female students.Improvements/Applications: It is required to develop a customized educational program to form a successful social relationship between adolescent female students.