摘要:Objectives: This study aimed to identify factors that affect the formation of a child's self-esteem from a holistic perspective through big data analysis. Methods: This study analyzed data from 560 children who participated in the tenth panel study of Korean children (third grade in elementary school) in 2017. The SPSS 20.0 program was used to organize variables, and the {mice} and {glmnet} packages in R 3.6.3 were used to handle missing values and explore predictors through Elastic net regression. Results: Among the 491 predictors, 19 were selected as important variables: time with family, leisure activities with family, time to help the family, frequency of family's money worries, mother's daily smoking volume, subject preference, homework time, the degree of sincerity in school life, creative classroom environment, peer attachment, time spent with friends, leisure activity time, weekend leisure activity time, travel experience, overall happiness, and use of slang. The variables that children subjectively respond to had more significant effects on children's self-esteem than the variables to which parents or teachers objectively respond. Conclusion: This study rediscovered important variables that have not been the focus of previous studies on the development of self-esteem in children. Based on the results of this study, we provided some practical suggestions for the development of positive self-esteem in children.
关键词:self-esteem;big data analysis;Elastic net;panel study of Korean children