Based on the large-scale survey data of residents' travel, this paper explores the behavior of travel modes of travellers in Xining,the typical corridor valley pattern city of plateau. The binary logit model of the commuters was constructed, and the individual and family attribute indicators of the travellers were used as independent variables, and the travel modes (cars, buses) adopted as the dependent variables. In this paper, the SPSS software is used to establish the commuter travel mode selection model of Xining residents. According to the SPSS output results, the influencing factors of travel behavior are analyzed from three aspects: model saliency, model calibration result and variable influence degree. The results show that the constructed model has strong statistical significance. When constructing the travel mode selection model, the personal characteristics and family characteristics of the traveler should be considered comprehensively, and the family car ownership has the most significant influence on the travel mode selection behavior. It can provide some reference for the study of traffic choice behavior of urban travelers with similar characteristics, and also provide a theoretical basis for the formulation of traffic policy.

CICTP