Introduction: Delay in diagnosis of bleeding can be due to underestimation of the actual amount of blood loss during delivery. Therefore, this research aimed to compare the efficacy of web-based, simulation-based, and conventional training on the accuracy of visual estimation of postpartum hemorrhage volume., Materials and Methods: This three-group randomized clinical trial study was performed on 105 midwifery students in Mashhad School of Nursing and Midwifery in 2013. The samples were selected by the convenience method and were randomly divided into three groups of web-based, simulation-based, and conventional training. The three groups participated before and 1 week after the training course in eight station practical tests, then, the students of the web-based group were trained on-line for 1 week, the students of the simulation-based group were trained in the Clinical Skills Centre for 4 h, and the students of the conventional group were trained for 4 h presentation by researchers. The data gathering tool was a demographic questionnaire designed by the researchers and objective structured clinical examination. Data were analyzed by software version 11.5., Results: The accuracy of visual estimation of postpartum hemorrhage volume after training increased significantly in the three groups at all stations (1, 2, 4, 5, 6 and 7 (P = 0.001), 8 (P = 0.027)) except station 3 (blood loss of 20 cc, P = 0.095), but the mean score of blood loss estimation after training did not significantly different between the three groups (P = 0.95)., Conclusion: Training increased the accuracy of estimation of postpartum hemorrhage, but no significant difference was found among the three training groups. We can use web-based training as a substitute or supplement of training along with two other more common simulation and conventional methods.