Based on the methods of evaluating debris flow hazard degree being multiple but difficult to come into being a system, a new Bayes discriminant analysis method to assess debris flow hazard degree is presented. After taking the mainly factors of debris flow hazard degree into considerationï¼the 10 factors related with topographyï¼geologyï¼hydrology and meteorology and human activities are chosen as the evaluating indexesï¼and the influence of dimension of the indexes are eliminated by the method of range. Thenï¼the weights of indexes are determined by preference ratio method. Finallyï¼the data of 20 debris flow projects are taken as the training and testing samples so as to build the model of Bayes discriminant analysis for evaluating the debris flow hazard degreeï¼and the resubstitution method is used to estimate the ratio of mistake-distinguish. Typical debris flow hazard degree in Yunnan-Guizhou plateau and Miyun county is evaluated by this trained modelï¼and it is showed that the results are consistent with the actual situation and the evaluation results of other methods.