Objective：To understand the status and identify the determinants of self-treatment of rural floating population in Beijing. Methods: The data were derived from the Fourth National Health Service Investigation of China, part of which was about rural floating population in Beijing. According to Andersen health services utilization model, Binary Logistic Regression model was used to analyze the influencing factors. Results: In the study, 51.7% of the research objects were inclined to self-treatment when they ever felt uncomfortable or sick in Beijing. The Binary Logistic Regression model suggested that, some factors from predisposing, enabling and need, were statistically significant determinants of utilization of self-treatment. Conclusion: There exists a certain percentage of self-treatment behavior for rural floating population in Beijing, and we should take some measures to promote the efficacy and safety of self-treatment.