Abstract: The objective of this paper is to provide an introduction to generalized linear mixed models and to compare of the efficiency of the GLMM and LM. The method of GLMM is described that allows genetic analyses and predicting breeding value for resistance traits. Binary and ordinal response traits are simulated in this study, and the design is mixed family of full-half sib. The logit and log linkage function are applied, respectively. The method of Fisher score is applied to calculate the parameters in generalized linear mixed models. The results showed that generalized linear mixed model can predict breeding value to a nicety. In addition, the method of GLMM has a great advantage in predicting the breeding values for resistance traits.