Abstract

The present study was conducted to develop neural network model on school drop out, with two primary steps in mind. They were : First step, to differentiate between risk and protective factors by applying the decision tree analysis on the data about individual, family, peer, school and community-related variable. Second step, to integrate risk and protective factors that would simulated on the multi-axis. According to the neural network model, the more risk factors is cumulated, the more rate of school drop out is increased. But in the case of protective factors, its tendency is not clear. The neural network model showed that risk and protective factor were simulated on the multi-axis would predict school drop out mainly by non-linearity function. Because the various sub-dimensions was involved, the neural network model was fit male and high school student samples very well. Tailored approach of school drop out is avaliable by applying neural network model. The future study on risk and protective factors model of school drop out is desirable to simultaneous consider patterns and duration of school drop out.