Abstract: Effective dose assessment can be given even long after the irradiation exposure event by the method of EPR (electron paramagnetic resonance) dosimetry, which has been developed into a lead-ing method for retrospective dosimetry of individual radiation exposure. The principle of EPR dosimetry is based on the fact that the concentration of radiation-induced radicals increases with irradiation dose. There is still some difficulty in the measurement of low dose at present, so further investigation on the effect fac-tors of the method is required in order to lower measure threshold, decrease error and optimize the method. Several major effect factors of measurement threshold and result uncertainty in EPR dosimetry have been analyzed and the corresponding correction methods for the factors were proposed.