HASTAC 2023 Conference

The heightened discourse around safety and sexual harassment in public spaces after the hypervisible #Nirbhaya sexual violence and murder case in Delhi and the subsequent Criminal Amendment Act, 2013 which criminalized sexual harassment in public spaces saw an increase in feminist activism around these issues. The need for safer public spaces, and access to justice mechanisms in cases of sexual harassment in public spaces led to the rise of initiatives like the toll-free police helpline to report crimes in various states (such as the Dial 112 facility in New Delhi), increased use of surveillance technologies like CCTV cameras by citizens’ groups, and activist interventions such as safety audits (Jagori and International 2010) and crowdsourcing applications (Liu 2019; “Safecity.in” n.d.; Seltzer and Mahmoudi 2013; Viswanath and Basu 2015). But what does “safety” mean in a deeply stratified society like India? What role can crowdsourcing play in this context to inform policy action on building safer public spaces? These are some of the key questions our research paper is concerned with, especially since crowdsourcing has grown to be one of the key modes utilized by activists to record women’s safety perceptions of public spaces and crime hotspots to inform gendered mobility choices. However, biases proliferate in the data pipeline (Eickhoff 2018; Marda and Narayan 2020; Olteanu et al. 2019; Sambasivan, Hutchinson, and Prabhakaran 2020; Sambasivan et al. 2021; Solymosi, Bowers, and Fujiyama 2018) especially in how the data is collected, who collects/reports the data, and how the data is interpreted and visualized - which in turn affect how the data is perceived and used broadly. For crowdsourced data to be used meaningfully for on-ground change, it is crucial to uncover the underlying heterogeneity of street safety perceptions based on socio-demographic characteristics of respondents.

Using critical feminist approaches to data (D’Ignazio and Klein 2020), our paper employs qualitative methods to identify the various facets including sociocultural, platform access, gendered mobility and safety factors which influence the collection and interpretation of crowdsourced street safety data in New Delhi, India. We conduct a systematic literature review and participatory photo mapping exercises combined with safety audits (Dennis et al. 2009; Jagori and International 2010; Natarajan 2016) with women in two low-income communities in Delhi. Further, we utilize feminist grounded theory analysis (Charmaz 2006) of structured interviews with key stakeholders working on the issues of gender, mobility, and safety. Interviewed stakeholders range from academics and community organizers to policy researchers and crowdsourcing platforms. Our analysis allows us to a) examine the complex discourse on the many meanings of safety, and the notions of surveillance as safety, b) to understand who is left out and whose voices are counted in conversations around safety in public spaces, and c) to offer insights into the potential and challenges of using crowdsourced data for imagining inclusive, safer cities. While the focus of our study is on urban women’s safety in public spaces through stakeholder interviews and participatory photo mapping in low-income communities, we recognize that the implications of data bias and gendered mobility go beyond the context of gender and class in urban cities. Some of our stakeholder interviews point to the layered complexities of rural-urban, class-caste, and other broader contexts which we hope to address in future research.