TL;DR
This paper investigates why women use bike-sharing services less than men, focusing on the VELIB system in Paris. By combining GIS analysis, travel demand data, on-site observations, user surveys, and social media analysis, the study identifies key barriers related to accessibility, safety, social constraints, and environmental factors that disproportionately affect women's adoption of bike-sharing. The findings aim to inform the design of more inclusive and equitable bike-sharing systems for the future.
What Problem Does This Paper Address?
Despite the growing popularity of bike-sharing systems as a sustainable urban mobility solution, there's a significant gender gap in their usage. This paper tackles the question of why women are less likely to use bike-sharing services compared to men. The authors highlight that while urban planning is increasingly focused on active transport, current systems often fail to adequately consider the unique needs and concerns of women, who may face different challenges and have distinct mobility patterns due to societal roles and safety perceptions. This disparity not only limits the potential of bike-sharing as a truly inclusive transport option but also hinders its effectiveness in promoting sustainable mobility for all.
Methods
The study employs a multi-faceted methodological approach, integrating several data analysis techniques:
- Geographic Information Systems (GIS): Used to analyze geolocated open datasets (land use, sociodemographics, mobility infrastructure) to identify and characterize docking stations based on their accessibility for women. This involved creating thematic maps and calculating indices like the Territorial Data Index (TDI), Socio-demographic Data Index (SDDI), and Mobility Data Index (MDI).
- Structured Proprietary Data Analysis: Examined VELIB's travel demand data, including rental statistics, trip distances, and user demographics, to understand usage patterns and correlate them with accessibility levels.
- Onsite Observations: Conducted at selected docking stations using a checklist focused on universal design indicators to assess infrastructure and surrounding context relevant to women's needs.
- User Satisfaction Index Questionnaires (UESI): Administered to collect data on users' experiences, perceptions, and expectations regarding the bike-sharing service, specifically focusing on identified "Fairness Characteristics" (FCs).
- Social Media Data Analysis (Twitter): Utilized to capture a broader range of user opinions and concerns, employing sentiment analysis and word shift graphs to identify gender-specific discourse around the VELIB service.
Datasets
The research utilized a combination of publicly available and proprietary datasets:
- Open Datasets:
- Copernicus Land Monitoring Service (Land Use)
- National Institute of Statistics (Sociodemographic data: population, age, gender, nationality)
- OpenStreetMap-Geofabrik DB (Points of Interest, Public Transport Services, Cycling Infrastructure)
- French public data platforms
- Proprietary Data:
- VELIB (Syndicat Mixte Autolib et Velib Métropole) travel demand data (number of bikes, rentals, trip details, unique users, female users).
- Survey Data:
- User Satisfaction Index Questionnaires (UESI) administered to 407 users.
- Social Media Data:
- Twitter data collected via API (December 2019 - June 2020), comprising 44,262 tweets from 8,414 users.
Key Insights and Ideas
- The "30% Problem": Women constitute only about 30% of VELIB users, highlighting a significant gender disparity that needs to be addressed.
- Beyond Infrastructure: A Holistic View: The study emphasizes that improving bike-sharing for women requires more than just adding more bike lanes. It necessitates a comprehensive approach that considers safety, social perceptions, and practical daily needs.
- The Interplay of Factors: Women's decision to use bike-sharing is influenced by a complex web of factors, including the availability of bikes, proximity to stations, the quality of cycle paths, personal safety perceptions, social stigma, and even weather conditions.
- Children and Shopping are Major Hurdles: The lack of child seats and adequate carrying capacity on bikes, coupled with societal expectations around childcare and shopping, significantly deters women from using bike-sharing services for these common trip types.
- Safety is Paramount: Perceptions of danger and insecurity, both while cycling and at docking stations (e.g., poor lighting, lack of security features), are major deterrents for women.
- Social Norms Matter: Cultural perceptions and the idea that cycling is for lower socioeconomic groups can act as a barrier, especially when combined with concerns about appearance (e.g., wearing skirts, heels).
- Twitter Reveals a Politicized Discourse: Social media analysis showed that while both genders discuss VELIB, women tend to focus more on issues of reliability and the practicalities of using shared bikes for commuting, while men often engage in more critical or politically charged conversations.
Key Results
- GIS Analysis Identified Key Stations: The GIS analysis successfully identified a shortlist of 20 diverse docking stations, categorized by high and low accessibility for women, forming the basis for further investigation.
- Travel Demand Data Correlates with Accessibility: The study found that the utilization/capacity ratio of bike-sharing services is influenced by the level of accessibility of docking stations. Notably, stations outside the city center showed a slightly lower female user ratio.
- Onsite Observations Highlighted Deficiencies: A majority of observed stations exhibited negative features related to spontaneity of access and separate cycling infrastructure. Half of the stations also had negative aspects concerning safe environments and personal safety.
- User Satisfaction Linked to Multiple Factors: User satisfaction was strongly influenced by the ease of booking and sign-up, station proximity, the perceived possibility of using services with children, and the availability of adequate lighting at stations.
- Twitter Data Showed Gendered Concerns: Women's tweets frequently mentioned reliability issues and the use of shared bikes for commuting, while men's tweets often focused on critical hashtags related to service failures.
Limitations and Future Work
- COVID-19 Impact: The authors acknowledge that the data collection period coincided with the COVID-19 pandemic, which may have influenced user behavior and potentially biased the results. However, they also note that this period saw a significant increase in cycling in Paris, making the survey timely.
- Twitter User Demographics: Sociodemographic characteristics of Twitter users were not directly available, and gender was inferred using a deep learning model, which has inherent limitations (e.g., not accounting for non-binary genders).
- Future Work: The researchers plan to apply advanced data analytics techniques, such as Analytic Hierarchy Process (AHP) and Machine Learning (Factor Analysis, Bayesian Networks), to develop a hierarchical model for designing parameters that enhance women's inclusion in bike-sharing systems. The ultimate goal is to use the collected data to define guidelines and policies for more gender-equitable bike-sharing services.
Summary updated