dkd

 Building safer public spaces: Exploring gender difference in the perception of safety in public space through urban design interventions

Concern around personal safety is a significant factor in constraining women’s access to and use of public space. One strategy presented to address this involves altering the design of built environments. However, tension and controversy surround these arguments in the literature on safety perception in public space. This study seeks to explore whether the presence of three design interventions commonly cited in the literature act to enhance or reduce perceptions of safety in public space, focusing on women’s experiences in particular. Three design interventions are examined: public toilets, solid walls, and graffiti removal. To test these interventions, an imagebased randomised control trial with 104 participants was conducted in 2018 in London (UK). A series of control and treatment images simulating each design intervention was viewed and ranked by participants according to perceived safety. The findings of this study suggest that: the presence of public toilets does not affect perceived safety; eliminating graffiti has a weak significant impact on perceived safety; and removing solid walls leads to significant improvements of perceptions of safety, with the effect being stronger for women. These results suggest that to maximise investment effectiveness, urban design and planning policies that aim to increase perceptions of safety should be evidence-based, and need to integrate a gendered perspective. The presented technique could support urban design processes by examining the safety-enhancement impact of proposed public space interventions prior to their being rolled out.

 1. Introduction The ability to safely utilise the urban public realm has profound implications for people’s wellbeing. A wealth of research (Anderson, Ruggeri, Steemers, & Huppert, 2017; Cattell, Dines, Gesler, & Curtis, 2008; Giddings, Charlton, & Horne, 2011) illustrates the benefits of public space utilisation including strengthened social inclusion, reinforced feelings of belonging and improved mental health. Studies also suggest that intensive use of public space can stimulate the local economy by boosting tourism and recreational activities and thereby increasing property values (CABE, 2004). Despite interest over recent decades in rejuvenating public spaces to stimulate their use – notably in Europe, the US and Latin America – concerns around inaccessibility and the exclusion of particular groups, among them women, from these spaces remain largely unaddressed (Cattell et al., 2008; Soraganvi, 2017). According to Ratnayake (2013), concern for personal safety not only has a detrimental psychological effect but also limits a person’s freedom and choice to move in a public space, thus reducing its use. Women as a group feature prominently in the literature on this subject, as it is argued that they face especially acute safety concerns when accessing public spaces and considering travel patterns. A large body of literature documents gendered differences in uses of and travel patterns in the public realm. Blobaum ¨ and Hunecke (2005) argue that women’s concerns of personal safety can result in their exclusion from outdoor sport and leisure activities, contributing to decreased use and a more muted city life. The perceived safety of a place may result in women altering or altogether cutting out travel to places perceived as dangerous (Koskela, 1997; Gargiulo et al., 2020). Concern for personal safety thus has often been found to preclude women from full and meaningful inclusion in public spaces, thus limiting opportunities to effectively reap the benefits to one’s wellbeing that come from accessing the public realm (Blobaum ¨ & Hunecke, 2005; Thynell, 2016; Valentine, 1990). One promoted strategy to tackle issues of perceived safety in public space is the design and planning of the urban environment (Blobaum ¨ & Hunecke, 2005; Dym´en & Ceccato, 2012; Harvey, Aultman-Hall, Hurley, & Troy, 2015; Gargiulo et al., 2020; Jorgensen, Ellis, & Ruddell, 2002). Valentine (1990) argues that, particularly for women, “public environment can have an influence on…perception of safety and hence on their willingness to use spaces and places” (p.301), and so careful design and planning of public spaces should foster their use. A great deal of research details the consideration of perceived safety when designing and planning public spaces. Harvey et al. (2015) for instance, show how street skeleton – greenery, alignments and continuity of buildings and building street ratios – all impact perceived safety in public space, while Soraganvi (2017) argues that women are particularly deterred by poorly designed urban public spaces such as those with poor lighting, empty lots, lack of public toilets or inadequate signage. While scholars argue for the importance of considering women’s perceived safety when designing a space (Jiang, Mak, Larsen, & Zhong, 2017), these concerns are often neglected in practice (Burgess, 2008; Greed, 2005; Parker, 2016). This is due in part to a lack of practical evaluation tools and causal empirical evidence on their effectiveness in helping urban planners and designers to operate. For example, a recent Systematic Reviews and Meta-Analyses shows that while a range of built environment interventions have been hypothesised to increase safety perceptions, few empirical studies have shown their effectiveness, resulting in low-quality evidence owing to “possible confounders [involved in] complex social interventions such as these” (Lorenc et al., 2013), p.2. There is little evidence that widely used investment strategies such as improving street lighting or installing closed-circuit television (CCTV) have a significant impact on increased perception of safety. In light of this evidence, we aim to assess the link between gender and perception of safety in public spaces while addressing some confounding factors seen in correlational studies1 . By doing this, we provide a research strategy to test ex-ante the effectiveness of design interventions to enhance perceived safety using photo-simulations. To do this, we designed an image-based, randomised control trial (RCT) technique implemented through a purpose-built platform. We collected data from 104 participants in London (UK) in 2018 that stated their perceived safety of a series of control and treatment images representing three urban planning interventions in a public space: building public toilets, removing solid walls and removing graffiti. Through this experiment, we generated reliable estimates of whether these urban interventions have an impact on perceived safety in public space and whether there are differences between women and men. The contribution of our paper is threefold. Firstly, we test the effectiveness of three widely promoted public space interventions on perceived safety on a sample of participants in London, providing evidence to promote designs that enhance safety perception in public space. Secondly, we present an explicitly gender-focused piece of work that adds evidence to the existing urban research on the relation of women’s safety perception to the built environment. Finally, by conducting this study, we present a RCT methodology based on photo-simulated scenarios that urban researchers and practitioners could use as a low-cost technique for testing the impact of urban interventions on the perceived safety of a public space before they are ruled out. In the next section, we present an overview of the relevant literature on public space and gendered perceptions of safety, detailing the three design interventions assessed in the study. This is followed by a methodology section covering the main considerations for the image-based RCT. We then present our findings with a focus on the gender difference in safety perception. Finally, we conclude with a discussion of the generalisability, limitations, and policy implications of this study. 2. Literature review 2.1. Public space: Gender inequality by planning and design There is increasing evidence that access to and participation in public space2 can offer a range of social, emotional, and psychological benefits (Anderson et al., 2017; Cattell et al., 2008; Garcia-Ramon, Ortiz, & Prats, 2004; Giddings et al., 2011; Mehta, 2014). Authors underline that the use of public space helps to build a sense of community by facilitating encounters and shared interactions, and promotes mental and physical health (Anderson et al., 2017; Cattell et al., 2008, Francis, Wood, Knuiman, & Giles-Corti, 2012; Gargiulo et al., 2020; Luymes & Tamminga, 1995). Studies have also linked the active use of and participation in public spaces with lower levels of exclusion and isolation. Sauter and Huettenmoser have even proposed that “public space is one of the crucial elements in any society for the social, cultural, economic and political inclusion of its members” (2008, p.68). Nevertheless, the benefits of using the public realm are not equally shared among members of society, as certain social groups have difficulty accessing or are explicitly excluded from public space due to their gender, age, ethnicity, abilities, economic backgrounds or citizenship status (Cattell et al., 2008; Garcia-Ramon et al., 2004; Jabareen, Eizenberg, & Hirsh, 2019; Williams, Logan, Zuo, Liberman, & Guikema, 2020). On the one hand, socioeconomic inequalities are built into spaces through design and planning. Low-quality public spaces tend to agglomerate in low-income areas, while socioeconomic spatial sorting means that vulnerable groups, such as female-headed households, are often found to “bear the brunt of poor environments” (Cavanagh, 1998, p.172; Williams et al., 2020). Since studies show that the quality of public space is related to the intensity of its use, vulnerable groups are disproportionately impacted by poor quality public space due to limited choices in where they live and work, and how they travel. (GarciaRamon et al., 2004; Jorgensen et al., 2002). On the other hand, exclusion from the public realm might also result from the lack of diversity consideration in urban planning. Theoretical and empirical studies show that poor urban design and planning of public spaces can serve as a powerful form of exclusion, deterring people from not only accessing public spaces but from navigating them with dignity and without fear 1 (Chiodi, 2016; Gargiulo et al., 2020; Luymes & Tamminga, 1995;). A vast body of literature recognises that design and planning do not take into account women’s uses of and experiences in public space, even though they differ considerably from those of men (Burgess, 2008; Greed, 2005). These differences can be illustrated, for instance, by comparing distinct mobility patterns in cities by gender. Feminist scholars argue that broader structural factors can disproportionately affect women’s movement patterns and experiences in cities around the world (Hanson, 2010; Beebeejaun, 2017; Chant, 2013). Women, for instance, are disproportionately responsible for caring and domestic duties and are overrepresented in part-time work. This results in women having more complex daily movement patterns and ultimately spending more time in public space than men (Levy, 2013; Garcia-Ramon et al., 2004; Whitzman, 2013). Generally, these considerations are not taken into account in urban planning, which instead bases its modelling methods on the male breadwinner’s daily trajectory of “home–work–home”. Current transport planning models thus find a failure to reflect “trip-chaining” travel patterns more common amongst women, such as “home–childcare–grocer–home–work–childcare–home” (Burgess, 2008; Greed & Johnson, 2014). Authors also argue that, due to having less access to private cars, women are more likely to walk or take public transport than men (Greed & Johnson, 2014; Whitzman, 2013). As a consequence, women may be exposed to encounters with strangers and therefore be significantly more concerned about safety when travelling (Thynell, 2016). These analyses underline how planning and design reinforce gender inequalities. Investigating women’s movements in and access to public space through the context of broader power structures highlights the need to consider and advance methods for the inclusion of women in the design and urban planning of public space (Cavanagh, 1998; Listerborn, 2016) 3 . 2.2. Women’s perception of safety in public space A key factor credited in shaping women’s movement in public spaces is fear of violence. Cavanagh (1998), p.169 argues that concern over personal safety profoundly shapes women’s lives and will “largely determine when and where women will go”. However, concerns of safety cannot be fully tackled simply by addressing crime – there is now robust evidence showing a lack of relationship between crime rates and perceived safety (Romer, Jamieson, & Aday 2003). A neighbourhood, for instance, can experience a reduction of crime over time while the perceived safety of its community remains unchanged. Notions of women’s safety thus move beyond merely the “absence of violence” to meaningfully integrating women’s perceived freedom and confidence in moving through and accessing public space without fear (Shaw, Andrew, & Whitzman, 2013). Studies have shown that women feel less safe than men in public space, displaying a range of hypotheses for this. The vulnerability hypothesis argues that perceived safety is the product of an individual’s conception of their own vulnerability to victimisation (Baur, 2007). Compared to men, women’s physical and social vulnerability results in a greater perception of risk in the built environment (Skogan & Maxfield, 1981). Indeed, studies show that women’s susceptibility to “harassment and (sexual) violence, including staring, groping, remarks and stalking as well as assault” (Whitzman, 2013), p.39 contribute to feeling less safe in public than men, regardless of reported crime rates (Pain, 1997; Warr, 1984). Conversely, the emotional gender-stereotypes hypothesis postulates that while women are encouraged to express their emotions, men are expected to repress these feelings, including fear (Sutton & Farrall, . Men are taught to play a protective role and to believe that physical strength can prevent them from becoming victims of crime. This can lead to a tendency to minimise their fear and the risk of victimisation in a public space (Tulloch, 2000). Existing correlational evidence explores the differing perspectives of men and women on the implementation of safety interventions in public space. For example, Yavuz and Welch (2010) indicate that the presence of cameras in public transit stations increases the perceived safety for both genders, with an impact notably higher in men. A UK study shows that while men can feel unsafe in the presence of groups of men, women can feel unsafe in the presence of a single man, with the reasons being gendered concerns of violence and sexual assault, respectively (Crime Concern, 2004). Research has also shown that women are more likely than men to avoid walking in dark public spaces, and to feel less safe in degraded or isolated spaces or from signs of uncivil behaviour such as vandalism or graffiti (Crime Concern, 2004; Yavuz & Welch, 2010; Gargiulo et al., 2020) and prefer to drive or take a taxi rather than walk (Atkins, 1989). This evidence suggests that policymakers should seriously consider a gendered perspective when planning public spaces if seeking to maximise perceived safety (Blobaum ¨ & Hunecke, 2005; Pain, 1997; Valentine, 1990). Despite the valuable work of the researchers noted above, such feminist perspectives remain peripheral to mainstream urban research and planning (Parker, 2016) while evidence remains correlational. As such, our study seeks to strengthen this body of work by offering causal evidence in support of the gendered nature of perceived safety in public space and practical methods to introduce a gendered lens in urban research and planning policy. 2.3. Focusing on interventions in the built environment Several theories describe the potential mechanisms through which the built environment could impact perceived safety, leading to various urban design and planning strategies to address concerns of safety. Here we assess the impact of three theory-driven interventions commonly debated in the literature (Austin & Sanders, 2007; Beebeejaun, 2017; Greed, 2016; Chiodi, 2016; Jacobs, 1961): (1) the provision of public toilets based on feminist theories of women’s restricted mobility, (2) the elimination of solid walls grounded in the “eyes on the street” theory of passive control, and (3) the removal of graffiti founded on the “broken windows” theory of stopping signs of disorder from spreading. 4 There are, however, many tensions and gaps in the literature surrounding these interventions, allowing for a valuable study to assess their impact on perceived safety. 2.3.1. Public toilets Accessible and adequate public toilet provision in public space has been underlined as a key gendered issue affecting women’s mobility and safety (Beebeejaun, 2017; Greed, 2016). Inadequate toilet provision can serve to restrict the mobility of women in the public realm, as they tend to require access to toilets more often than men. Studies show that women generally urinate more frequently and may require increased toilet use for reasons related to menstruation, pregnancy or menopause (Greed, 2016), in addition to their prevalence as carers for vulnerable groups such as the elderly, children and people with disabilities (Afacan & Gurel, 2015). Research also shows that women are more frequent public transport users, often accompanying children or elderly family members (ibid.). As a result, a lack of provision in quantity and quality of toilets may restrict choices of routes and destinations for women. Increased attention is being drawn to the link between public toilet provision and women’s safety. In 2011, United Nations Women (UNWomen) argued that women’s safety is compromised when female public toilets are situated in covert, poorly lit areas, and when men’s public toilets open directly onto sidewalks (UN-Women, 2011). This report states that “better situated and maintained public toilets can go a long way in preventing sexual harassment of women” (p.7). It can thus be argued that adequate toilet provision and design help to increase women’s perceived safety and, therefore, their participation in the public realm (Afacan & Gurel, 2015). We seek to gain greater insight into how public toilets impact perceived safety, particularly for women. 2.3.2. Solid walls A common strategy to promote safety is to establish an urban environment that ensures greater visibility of the public space, or allows for more “eyes on the street” (Chiodi, 2016). The frequently cited “eyes on the street” theory originates from sociologist and urban planner Jane Jacobs who suggests that more people, or “eyes,” serve as a form of informal surveillance. By extension, it is argued that this increases the likelihood of witnesses and bystander intervention, thereby reducing crime and increasing perception of safety (Chiodi, 2016; Jacobs, 1961; Sweet & Ortiz Escalante, 2010; Scott et al., 2008). Interventions to promote more “eyes on the street” have included the strategic placement of windows and entrances, the redesign of green spaces and the removal of solid walls or similarly large obstructions to public space visibility (Chiodi, 2016; Cozens, Love, & Nasar, 2015). UN-Women (2011) also cited improved visibility, informal surveillance and “eyes on the street” as essential strategies to establish safer cities for women. Despite its popularity, Jacobs’ theory and similar approaches have received criticism, mainly due to a lack of causal evidence on the reduction of crime rates or improved safety perception through purposeful design (Anderson et al., 2013). Notwithstanding these critiques, correlational research and safety audits have demonstrated that people, particularly women, tend to feel less safe in closed-off, isolated areas, and often feel safer in well-maintained spaces that have more people, higher levels of activity and non-enclosed vegetation settings and are in streets lined with houses, shops, restaurants and windows (Cattell et al., 2008; Gargiulo et al., 2020; Harvey et al., 2015; Jorgensen et al., 2002; Mehta, 2014; Sandberg & Ronnblom, ¨ 2015). We seek to add to causal evidence in determining whether or not improved visibility on the streets does indeed influence perceptions of safety in urban public space. 2.3.3. Graffiti A number of studies have argued that graffiti can lower perceived safety by increasing fear of crime (Austin & Sanders, 2007; Mehta, 2014; Rader et al., 2012). These arguments draw from Wilson and Kelling’s “broken windows” theory, which posits that vandalised, neglected and poorly maintained areas provide clues to criminal activity in the area and show that residents care little about the social or physical condition of their neighbourhood (Wilson & Kelling, 1982). Graffiti, in this context, is argued to indicate neighbourhood incivilities and a neglected urban environment, enhancing public fear and, therefore, concerns of safety (Adu-Mireku, 2002). Similarly, some gender-sensitive urban planning research has suggested that removing graffiti will specifically improve women’s perceptions of safety (UN-Women, 2011; Valentine, 1990). “Tough on graffiti” policy approaches have thus been popular amongst many local governments, particularly in the UK (Vanderveen & van Eijk, 2016), pp.109. Nevertheless, researchers have questioned this approach and its theoretical underpinnings, finding more nuance in the public’s view of graffiti. Research has illustrated how people’s perceptions of graffiti can differ vastly. It may be deemed as offensive or fear-inducing, welcomed as a form of art or condemned as a crime (Shobe & Banis, 2014). Indeed, graffiti consists of a variety of forms, ranging from smaller tags to largescale street art, and these different forms may be regarded and valued differently (Austin & Sanders, 2007; Vanderveen & van Eijk, 2016). We explore whether removing graffiti, specifically tags, from the urban landscape could be considered a useful strategy to help promote perceived safety, particularly among women. 2.4. Photosimulation and perceptions of safety in the built environment Photo simulations, a technique that involves manipulating photographs to recreate changes to the built environment, has been widely used to understand people’s preferences in urban planning and environmental psychology research (Jorgensen et al., 2002; Kuo, Bacaicoa, & Sullivan, 1998; Rodiek & Fried, 2005). It has been often used to study the impact of landscape changes, for example, the effect of river restoration on the satisfaction of individuals’ needs (Junker & Buchecker, 2008), the influence of architectural design on adult’s satisfaction with care home facilities (Cerina, Fornara, & Manca, 2017), or the effect of including car parks and natural elements over residents’ preferences of commercial strip developments (Sullivan & Lovell, 2006). A specific body of research uses photo-simulations to study the impact of built environment transformations on perceived safety. A number of studies focus on the effect of different types of green infrastructure on perceptions of safety, finding positive effects (Jiang et al., 2017, Kuo et al., 1998, Navarrete-Hernandez & Laffan 2019). A few studies have specifically utilized photo-simulation to assess urban environment transformations’ impact on gendered perceptions of safety. One of these exceptions is Jiang et al. (2017) whose study simulates improvement of alleyways, showing that assigning both urban functions (e.g. cafes or bike parking) and vegetation contributes to closing the safety perception gap between men and women. Jorgensen et al. (2013) use the presence or absence of people in photos of public parks to show that perceptions of safety increase at a higher rate for men than women. This body of literature demonstrates that photo simulation can be used to understand people’s preferences for urban environments and, particularly, perceptions of safety. In this study, we use photo simulation to inform the design of public spaces that enhance perceptions of safety, notably for further inclusion of women in public space. 3. Empirical strategy 3.1. Study design To test the impact of these design interventions on perceived safety, we conducted an experiment to compare participants’ perceptions of different urban spaces using computerised photo-simulations. We used an online platform, Urban Experiment (www.urban-experiment.com), to collect a data sample of 104 individuals walking in open streets in the London School of Economics (LSE) campus in 2018. Participants were presented with a series of six randomly assigned public space images, either control or treatment (with or without the incorporation of a safety-enhancing urban design or planning intervention, respectively), then rated them according to how safe they would feel walking alone in the presented public space. As with any RCT, this means that our results have strong internal validity, however, they should not be extrapolated to the larger population of London residents.