Where the station ends and surroundings begin: A holistic gender perspective on perceived safety in public transport

Abstract

Women and gender non-conforming individuals (GNC) often feel less safe at and around public transport (PT) stations, compared to men. The built environment (BE) plays a well-documented role in shaping perceived safety, yet it remains unclear which specific features are most responsive to the safety concerns of women and GNC. Using data from a tailor-made online survey (N = 3101) from 232 train stations in East Denmark, we investigate BE features explaining each gender group’s perceived safety at the home-end and activity-end travel environments. A structural equation model with multi-group analysis (SEM-MGA) shows that well-lit travel environments and clear wayfinding improve perceived safety for all groups, whereas isolated areas, parking lots, and closed facades around stations particularly harm women’s and GNC travellers’ perceived safety. There are also slight differences between trip ends: shops, urban life, and trees have a stronger positive impact at the home end, while closed facades and isolated areas more strongly reduce perceived safety at the activity end. Towards reducing the gender gap in perceived safety, and achieving a more inclusive PT system, policymakers should prioritise gender-sensitive approaches when designing stations and their urban surroundings. • Women and gender non-conforming (GNC) travellers’ perceived safety in PT requires attention. • SEM-MGA reveals gendered differences in built environment effects on perceived safety. • Well-lit environments and clear wayfinding enhance safety for all groups. • Isolated areas, parking lots, and closed facades especially harm women & GNC safety. • Significance of the built environment differs across home and activity ends of the journey.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 3 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.