In this study, prediction models were constructed for four subfactors (self-understanding and sociality, job understanding, career exploration, and career design and preparation) of high school student’s career development competencies using data from the Korea Career Education Survey 2019. In addition, considering that the machine learning methods used in the previous studies were limited in providing an explanation, 20 predictors with a high contribution for each subfactor were evaluated using SHAP (SHapley Additive exPlanation), an explainable artificial intelligence. The main results are as follows. First, overall school life and career activity satisfaction showed a high contribution in all subfactors. Second, school career activities highly contributed to the prediction of high school student's career development, and the contributions of individual career activities differed by the subfactors. Third, regarding the career education curriculum, it was found that the contribution of the number of career activities in creative experiential activities was high. Fourth, among the student variables, conversation with parents showed a high contribution in all subfactors. Fifth, for all subfactors, the use of career programs and materials and whether it was a career education site did not appear. In addition, in comparing the individuals with the identical predicted value, it was found that the contribution and direction of variables for each student were different. Based on these main results, discussions on career education and programs are presented.