A clustering analysis of car-hailing travel behavior based on latent class model: a study from a prefecture-level city in China

This article used a latent category regression model to explore how different factors affect the travel behavior of car-hailing users. Online and offline surveys were distributed across various districts of Pingdingshan, Henan Province, resulting in 527 valid responses. Data analysis was done using Stata16 and RStudio software for descriptive statistics and regression analysis. Two user groups were identified: opportunistic passengers and loyal passengers. Waiting time and travel environment emerged as key factors in this classification. Longer waiting times led to lower overall user satisfaction. Comparative analysis showed that the travel environment had a significant impact on opportunistic passengers, surpassing its effect on loyal passengers with a 95% confidence level. In newly developed urban areas, opportunistic passengers were 42% more likely to recommend car-hailing services. Additionally, higher passenger satisfaction was linked to increased car-hailing usage, rising from less than weekly to at least six times weekly. These findings offer valuable insights for improving urban transportation infrastructure.