Objective To evaluate the uncertainty in the determination of lead in tea by inductively coupled plasma-mass spectrometry (ICP-MS) and to improve the method so as to enhance the quality of the analytical results.Methods Based on the mathematical model and the experimental procedure, the sources of uncertainty components were analyzed and combined.Results When the lead content in tea was 0.346 mg/kg, the expanded uncertainty was 0.026 mg/kg at a 95% confidence level (k=2). After method improvement, the combined relative standard uncertainty decreased from 0.038 to 0.019, and the expanded uncertainty decreased from 0.026 mg/kg to 0.013 mg/kg. The measurement result was (0.341±0.013) mg/kg (k=2), indicating a significant reduction in uncertainty.Conclusion Calibration curve fitting and sample weighing are the main sources of uncertainty in this experiment.