Study On Travel Mode Choice Behavior Of College Students Based On A SEM-LOGIT Model

With the rapid development of social economy and the adjustment of policy orientation,ride-hailing,as a representative of emerging shared travel services,has been rapidly popularized in cities and has a profound impact on people’s travel.As an important part of young people,college students’ travel behavior has a significant impact on urban traffic.The study of College Students’ travel mode choice behavior plays an important role in the continuous guidance of China’s future transportation policy.Using the ride-hailing travel data of college students in 2019,this paper constructs a structural equation model for the comparison between ride-hailing and taxi,explores the relationship between positive and negative emotion,operation service,satisfaction and loyalty,and compares and analyzes the key factors affecting college students’ travel mode choice behavior by constructing MNL model,RPL model and SEM-RPL model.According to the results of SEM-RPL model,the marginal effect analysis is carried out to further explore the impact on various travel modes under the changes of ride-hailing and taxi travel time,travel cost and loyalty,and put forward targeted policy suggestions.The main conclusions are as follows:1.In terms of ride-hailing and taxi,the impact of operation service,positive emotion and negative emotion on passenger satisfaction decreases in turn.Positive emotion and operation service have a positive correlation with passenger satisfaction,negative emotion has a negative effect on passenger satisfaction,and passenger satisfaction has a positive and significant impact on loyalty.2.The increase of travel time and travel cost will reduce the utility of college students choosing various travel modes.With the increase of travel number and travel distance,taxi is more favored by college students than public transport.ride-hailing discount and premium strategy will significantly affect college students’ willingness to choose ride-hailing.In addition to grade factors,other personal attribute factors have a significant impact on travel behavior.3.MNL model and RPL model show differences in travel costs,while RPL model captures heterogeneity through travel costs.Compared with RPL model,SEM-RPL model can better explain college students’ travel mode choice behavior,and latent variable loyalty significantly affects college students’ choice of ride-hailing and taxi.And put forward policy suggestions for the development of ride-hailing and taxi companies through marginal effect analysis.