Academic resilience is evident in students who are living in vulnerable environments, yet achieve success in academic outcomes. As a result, substantial attention has been devoted to identifying the factors associated with academic resilience and supporting students to be resilient. This study used the Classification and Regression Tree and Multilevel Logistic Regression modeling to identify the potential factors related to students' academic resilience. Using these tools, the study analyzed the B-S-J-G (China) sample in PISA 2015. The variables that significantly predicted whether a student is disadvantaged and resilient (DRS) or not resilient (DNRS) were shown to be: Proportion of teachers in school with master's degrees, Proportion of teachers in school with bachelor's degrees, Environmental awareness, Science learning time per week, Number of learning domains with additional instruction, and Students' expected occupational status. These findings may enlighten governments, teachers, and parents on ways to assist students to be resilient.