Exploring Ecological Factors Associated with At-Risk Students with Suicidal Ideation: A Decision Tree Algorithm

Abstract

The primary purpose of this study was to explore the ecological factors of at-risk students with suicidal thoughts among second-year middle school students, participated in the 5th year of the Gyeonggi Education Panel Study (GEPS). The decision-tree model, one of the machine learning algorithms, confirmed the intersectionality between suicidal ideation and the ecological factors of students at risk. For students who answered “strongly agree” and “not at all” to the question of suicidal ideation, mental health, attachment alienation, academic stress, gender, household income, and delinquency factors were identified as major factors with statistical significance. Based on the decision-tree model and its results, it was emphasized that it is necessary to understand the psychological and emotional needs, the home environment, and the school environment of students in crisis from an ecological perspective. In addition, the implications of the data-based approach, such as the decision tree model for School-Wide Positive Behavior Supports and the school safety integrated system (Wee project), were discussed.