Differences in street-scale built environment preferences towards biking: a latent class analysis of stated choice data

Liu, Yanan; Yang, Dujuan; Timmermans, Harry J.P.; de Vries, Bauke · 2021 · DOAJ

DOI: 10.1080/19463138.2021.2004545

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates how street-scale built environment attributes influence cyclists’ preferences for biking as an access/egress mode to metro stations, specifically addressing the under-researched context of emerging economies. The research is motivated by the shift in Chinese urban planning toward automobile-centric infrastructure, which has degraded cycling conditions and hindered the integration of biking with public transit. By understanding preference heterogeneity among cyclists, the study aims to inform the design of streets around public transportation hubs to encourage sustainable mobility. The researchers employed a stated choice experiment conducted in Tianjin, China, focusing on the Yingkoudao metro station area. Data were collected via face-to-face interviews with 803 respondents in September 2018. The experimental design systematically varied eight street-scale attributes: road segment length, average building height, retail shop frontage, cycling facilities at intersections, bike lane width, street-side greenery, crowdedness, and lamp density. Each attribute was defined at four levels, and respondents completed four choice tasks selecting between two street profiles or a "none of these" option. To analyze preference heterogeneity, the authors estimated a latent class model using Nlogit 5, incorporating individual socio-demographic characteristics (age, gender, education, driving ability, and car ownership) into the class membership function. The analysis identified two distinct latent classes of cyclists. Class 1 comprised 78% of the sample and exhibited higher general utility for biking as an access mode. Class 2, representing the remaining 22%, derived lower utilities from street-scale attributes, indicating a lower preference for biking to the metro. Both classes strongly preferred streets with both traffic lights and fences at intersections, wider bike lanes, and low vehicle/cyclist crowdedness. However, significant differences emerged in other preferences: Class 1 preferred shorter road segments and low-rise buildings (one to three floors), whereas Class 2 preferred medium-height buildings (four to six floors) and showed a positive marginal utility for shorter road segments. Regarding retail frontage, Class 1 favored streets with some (25%) or half (50%) shop coverage, while Class 2 showed no significant deviation from the mean for retail levels. Class 1 preferred high-density greenery (trees and hedges) and moderate lamp spacing (15–30 meters), while Class 2 preferred either trees or hedges and wider lamp intervals (>30 meters). The findings demonstrate that cyclists are not a homogeneous group regarding street design preferences, particularly in the context of public transit access. The existence of distinct preference clusters suggests that urban planners should consider targeted infrastructure improvements rather than one-size-fits-all solutions. Specifically, ensuring safety through intersection facilities and dedicated bike lanes is universally important, while aesthetic and structural elements like building height and greenery density may appeal differently to various user segments. These insights are critical for designing inclusive, cyclist-friendly environments around metro stations in rapidly urbanizing cities.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• urban rural setting