Research On Taxi Service Mode Choice Behavior Based On Theory Of Planned Behavior

In recent years,with the development of Internet technology,a new kind taxi named Ridesourcing which can take on the Internet has appeared,such as Didi and Shen Zhou etc.The emergence of Ridesourcing can effectively solve the asymmetric information between passengers and drivers and fill the taxi ownership vacancy.Reasonable price,high quality,convenient service and rapid response make Ridesourcing gradually become a popular way to travel.Ridesourcing has become a new mode of residential travel.First of all,this paper compares and analyzes the characteristics of the traditional taxi and Ridesourcing,and makes clear the similarities and differences between different taxi service modes;At the same time,based on the analysis of the travel demand hierarchy and psychological state of the choice of taxi service model,improves the perceived quality of service quality,and improves the traditional theory of planned behavior.Designed questionnaires to collect data,using the reliability and validity of SPSS test data,and made descriptive statistical analysis of the data.Secondly,using the data after the inspection,this paper constructs two models.The first one is the structural equation selection model,this paper examines and modifies the hypothetical model of taxi and Ridesourcing two taxi service modes,and determines the selection model of different service modes.Proved that the two way taxi passengers "perceived behavior control" to "behavioral intention" has a significant positive influence and the traveler’s preference for different taxi modes of travel has no significant influence on the behavior intention of the mode choice.The second one is the mixed model combined SEM and Logit,using structural equation model to quantify the theory of planned behavior variables.Then adds those variables to Logit model,calibrates the Logit model without considering TPB variables and the Logit model considering TPB variables by SPSS,and compares the accuracy of the two methods.It proved that the model precision considering TPB variables is higher than that without latent variables,and the estimation results of TPB variables are consistent with those obtained in SEM model.Finally,according to the analysis of the model results in the two part,this paper provides a theoretical basis for the formulation and improvement of the relevant policies and measures of the traditional taxi market and the emerging ridesourcing market from the perspective of passenger decision-making behavior.