Examining user attitudes towards ride-hailing services – A SEM-MIMIC Ordered Probit approach

Ride-hailing services (RHS) are rapidly transforming the urban transportation landscape, and subsequently, users’ perception of mobility services. Hence, it is of utmost importance to understand the perceptual and latent attitudinal factors that drive such service usage. This gains more relevance in the context of a developing nation both because of its characteristically different transport environment- lifestyle interaction, and relatively fewer studies investigating RHS utilization. The present research uses revealed preference data from household surveys (N = 418) to estimate the usage propensity of RHS services in Kolkata, as it has the highest share of commuters among the four large metropolitan areas in India. A SEM-MIMIC Ordered Probit modelling framework has been developed, as it extracts the advantages of exploring latent constructs through a structural equation model (SEM) and examines their interaction with demographics and trip-specific factors with the Multiple Indicator Multiple Cause Model (MIMIC). This study relies on the confirmatory approach to establish the latent attitudinal factors which stem from accepted theories in travel behaviour, i.e., Theory of Planned Behaviour (TPB) and Technology Acceptance Model (TAM). Subsequently, the Ordered Probit model estimates RHS use-frequency. The results highlight that the latent variables (LVs), viz., ride-hailing attitude and perceived usefulness, are the most significant while estimating RHS utilization. These findings encourage RHS providers to focus on service aspects (for example, security during rides, clean vehicles, reliability of rides, less wait time between booking and ride arrival), instead of only emphasizing on (cosmetic) changes to the booking interface. Besides, in contrast to the developed countries, subjective norms were found to have an inverse relationship with RHS usage, suggesting inhibition among the public, which is probably arising from the dearth of customer-friendly service, especially after being comparatively expensive. The model also suggests the supplementary role of RHS to public transit, which could be pivotal in its integration into mobility-as-a-service (MaaS) and also calls for regulatory actions. The demographics (e.g., age, gender, household income) and trip-specific (e.g., trip purpose, trip length, time-of-day) covariates add further meaning to the relationships among latent constructs. The results suggest a higher preference for RHS among non-car-owners, whereas frequent use of ride-hailing is observed to have a likely positive association with longer trip lengths. Overall, this research brings valuable and first-of-its-kind insights into attitudinal factors and their interaction with demographics and trip-specific covariates facilitating RHS utilization in the context of a developing nation.