Original Article Shared space : design, user perception and performance

Borja Ruiz-Apila´neza,*, Kayvan Karimib, Irene Garcı´a-Camachac and Rau´ l Martı´nc aDepartment of Civil Engineering and Built Environment, School of Architecture, University of Castilla-La Mancha, Escuela de Arquitectura, Avda. Carlos III, s/n, 45071 Toledo, Spain. E-mail: [email protected] bFaculty of the Built Environment, The Bartlett School of Architecture, University College, London, UK. cDepartment of Mathematics, School of Architecture, University of Castilla-La Mancha, Toledo, Spain. *Corresponding author.

Abstract Shared space is becoming an increasingly well-accepted approach to design, pursuing improvement of street uses on foot without restricting other modes of movement. This approach introduces a degree of uncertainty regarding application of rules by minimising traffic signs and conventional physical demarcations to enable different users to share the space and negotiate their movement through and across it. Despite existing evidence for reduced traffic accidents and casualties, this type of street design remains an unorthodox approach. This article addresses the key issues of road safety, user comfort and revitalization aspects of the shared space, raised in the last decade with a focus on pedestrian users. Based on analysis of the layout, performance and user perception of six study areas, evidence reveals how these schemes affect safety, comfort, the conventional spatial distribution or hierarchy of users, and public life. International (2017). doi:10.1057/s41289-016-0036-2

Keywords: shared space; street design; road safety; pedestrian comfort; street vitality; public space

Introduction InterregIIIprojectofthesamename (2005, 2008a, b), which was stimulated by the Shared space is becoming a widespread yet con- ideas of Hans Monderman and his early experi- troversial approach to street and public space ences in The Netherlands. The early diffusion of design that has attracted a great deal of attention this concept in the UK corresponds to one of the during the last decade among urban designers and project’s partners and main advocates, Hamil- those interested in improving the built environ- ton-Baillie (2004, 2008a, b, 2014). ment by rebalancing the conventional transporta- Similarly to the term ‘‘shared space’’, different tion modal share and enhancing pedestrian terminologies, such as , encounter zone, mobility. Controversies include the appropriate- home zone, shared street and shared zone, have ness of the term itself, the origin of its denomina- been used for different street layouts that do not tion, its definition and more importantly concerns segregate pedestrians from vehicles. These defini- regarding the performance and impact on users of tions have similarities and differences regarding this design approach. their objectives and design features, as broadly The term ‘‘shared space’’ has been widely used discussed by Karndacharuk et al (2014a). Accord- in literature since 2000, as well as in media and ing to Hass-Klau (1990), the very same term among city officials—mainly, but not only, in the ‘‘shared space’’ was already used in the UK during European context. The current proliferation of the 1960s and 1970s, referring to a traffic-calming the term originated with the 2005–2008 street layout applied to cul-de-sac residential

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International www.palgrave.com/journals Ruiz-Apila´nez et al streets. However, the term remained unused as a qualitative report prepared for the Department street design concept for decades, and it is only for Transport by MVA Consultancy (2011) showed recently that it has been re-coined with a related that 4 out of 20 blind or visually impaired users but different meaning. would not use a shared space street again, based Probably because the term was purposely not on an experience that took place across seven clearly defined by those responsible for its rebirth different sites in the UK. Quite recently, Havik et al and dissemination, currently there are a number of (2015) confirmed that blind users and those with different definitions available: ‘‘the meaning […]is evaluated shared space evolving as experience is gained and knowledge schemes more negatively, based on the experience expands’’ (Shared Space et al, 2008, p. 3). Despite of 25 people in two shared space locations and two more recent official definitions, such as the one by conventionally designed settings in The the Department for Transport (2011), which would Netherlands. apply only in the UK, the term in this paper is Imrie (2012, p. 2274) argued that, for many intended to refer to a set of built environment visually impaired people, shared space is a design principles that prefer to ‘‘combine rather disabling design ‘‘enlarging socio-spatial divi- than separate the various functions of public sions and inequalities in the urban environment’’. spaces’’ and seek to ‘‘improve the quality of public These concerns were questioned by Parkin and spaces and the living environment for people, Smithies (2012,p.135),whoultimatelysuggested without needing to restrict or banish motorized that ‘‘shared space needs to preserve a safe area traffic’’, as originally proposed in the EU project for pedestrians’’ and ‘‘provide a rich physical Shared Space (2005, p. 5). environment of contrasts in terms of surface On top of the above-mentioned issues, there are tactility, colour contrast, and the enhancement of others related to the functioning and impact of the sound and other sensory clues’’, stressing the call shared space approach, including the traffic safety from Methorst et al (2007)forsafeareaswithin of shared space schemes, how comfortable users shared space streets. As a result, tactile and feel sharing the space and their claimed impact on demarcatedareasthatarenotaccessibleto public life. These concerns are the main purpose of vehicles, i.e. safe zones, and other contributing this investigation, since they remain unclear after solutions have emerged in the most recent shared being partially addressed in previous studies, as space schemes. further explained in the next section. However, safety and comfort issues that are particularly relevant to particular groups of more vulnerable users are not the focus of this research. Research Objectives Rather, this investigation intends to focus on the most vulnerable users of shared space streets Since the early experiences in The Netherlands, from a broader perspective, i.e. that of an average shared space schemes are reported to have effec- pedestrian. Among the possible conflicts between tively reduced traffic incidents and accidents users, the focus here is placed on the pedestrian– (Commission for Architecture and the Built Envi- driver interaction, given that the share of cyclists ronment, 2007; Euser, 2006; Hamilton-Baillie, found in the analysed study areas (SAs) is 2008a; Karndacharuk et al, 2014b). Despite the negligible. Therefore, the main purpose of this lack of sufficient published research on this topic, research is to investigate how safe and comfort- the greatest criticisms do not question the objective able pedestrians feel in shared space streets. As safety of the schemes, i.e. reduction in casualties, has been advanced and is further discussed but focus mainly on the subjective or perceived below, recent research by Hammond and Mus- safety and comfort experienced by users. Among selwhite (2013), Kaparias et al (2012) and Moody others, Methorst et al (2007) argued that the and Melia (2014) investigated these issues, but decrease in accident/casualty numbers could have their results remained inconclusive and/or con- been created by an overall reduction of pedestri- tradictory, motivating the type of research pre- ans, caused by their discomfort with the imple- sented herein. mented schemes. In addition, this study tackles another important This issue of perceived unsafety and discomfort question that has also remained inconclusive to is emphasised for the case of blind or visually date: Does shared space enhance public life? This impaired users, according to the Guide Dogs for revitalization potential is claimed by advocates, the Blind Association (GDBA, 2006). The who claim it can foster ‘‘multiple uses of streets

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Shared space streets and spaces for every kind of social activity’’ peculiarities and variability present along Exhibi- (Hamilton-Baillie, 2008a, p. 137), and governmen- tion Road, two control streets were included in the tal publications, which refer to it as a ‘‘way of present research as well. The specifics are further enhancing a street’s sense of place [the quality described in the section on the SAs. which makes a street somewhere to visit and spend time in]’’ (Department for Transport, 2011, p. 6). Recent research by Biddulph (2012a, b), Curl Observations et al (2015), Karndacharuk et al (2013), Hammond and Musselwhite (2013) and Moody and Melia A set of systematic on-street observations were (2014) is, again, often inconclusive and/or contra- performed, including counting, mapping and trac- dictory, making it pertinent for new research to ing mainly pedestrian activity. A previous pilot investigate such social issues further. This study study by the authors and the reports available from carefully takes into consideration the work by the the Royal Borough of Kensington and Chelsea above-cited researchers and builds on it to develop (RBKC) on Exhibition Road (MVA Consultancy, an applicable set of research methods. 2012, 2013a, b;SYSTRA,2014) confirmed that lunch hours were the most active daily period in terms of pedestrian activity. Accordingly, a series of system- Methodology atic observations were carried out from 12:00 to 14:00, on both autumn weekends and weekdays, on This research analyses and compares six different 23 and 27 November 2013, with no rain and usual SAs, combining a set of different methods, includ- London weather conditions for that time of year ing quantitative and qualitative descriptions of (partially cloudy and temperature around 10 °C street design and adjacent ground frontages and during lunchtime). The following observations uses, as well as formal and informal observations were included in this study: of human activity to assess performance. Further- (a) Vehicular and pedestrian hourly volumes were more, an on-street user survey was used to counted for the different street sections based investigate pedestrians’ attitudes towards the on two 10-min counts per hour, noting the shared space street schemes. Differently from the number of pedestrians walking along the road; majority of previous studies, drivers were not (b) Pedestrians crossing the street at each section considered in this study, to focus more on pedes- were counted and mapped for two 10-min trians as more vulnerable users. intervals to evaluate freedom of movement and the assumed priority for pedestrians when crossing the carriageway; Comparing streets (c) The location of people standing or sitting, the only stationary activities observed in the area, Similarly to the classic street liveability study by was also recorded twice from 12:00 to 14:00 at the Appleyard et al (1981), the methodology applied in six street sections to provide objective and this work compared six sections of Exhibition Road - comparable data regarding on-street social activ- and two other streets in South Kensigton, London, ity. For sitting, the observations distinguished which offer different street layout treatments. Four between people using primary seating (public different sections of Exhibition Road, a recently chairs and benches), secondary seating (, refurbished street, account for similar but still differ- stairs, low walls and other public elements) and ent examples of shared space layout. Two sections of tertiary seating (provided by cafes, restaurants two different streets nearby, Queen’s Gate and etc.), as suggested by Ruiz-Apila´nez et al (2014). Cromwell Road, were examples of more conven- tional streets with kerbed footways, showing some similarities and differences in terms of adjacent land use, road width, footway width, vehicular traffic Survey questionnaire flow, available sitting spaces and amount of trees. These latter two were used as a control group, The questionnaire was designed to evaluate peo- as suggested by Hammond and Musselwhite ple’s perception of and attitudes towards the (2013, p. 95). Havik et al (2015) applied a similar different street layouts, taking into account previous approach, comparing two shared space streets research on shared space with similar objectives, with two conventional streets. Given the namely the studies of Kaparias et al (2012), which

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Ruiz-Apila´nez et al used web-based questionnaires about virtual, ver- samples originated from the same distribution for bally described street scenarios, Hammond and two groups or more than two groups; Kendall’s Musselwhite (2013), who performed 100 on-street tau-c test was employed for identifying correlation questionnaires for Widemarsh Street (, between ordinal data. These provided more confi- UK), and Moody and Melia (2014)onElwickSquare dence in results given the non-normality of the (Ashford, UK), which included 144 semi-structured answer distribution to questions 2–4. A significance interviews. level of 0.05 was considered for all tests. The The survey employed in this research includes median was considered the most representative 305 people who fully answered the following measure of central tendency, and box-plots graphs questions: are used to present the sample. The results and specific tests employed to investigate the outcomes (1) Do you live or work in the area? (a) Yes, (b) no. on different topics are discussed throughout the (2) In a scale from 0 to 5, how safe do you feel general discussion of this article. using this street in relation to traffic? [A Considering the survey sample, 309 surveys subjective slider from 0 (not safe at all) to 5 were started and 305 were completed. Only the (very safe) was available for the respondents]. answers of the latter were considered in the (3) In a scale from 0 to 5, how comfortable do you analysis. Respondents were randomly picked but feel using this street in relation to traffic? [A uniformly distributed among the six case examples. subjective slider from 0 (not comfortable at all) The percentages of completed interviews were to 5 (very comfortable) was available for the equally distributed across the SAs, ranging from respondents]. 16.2 to 16.8 per cent. The majority of respondents (4) In a scale from 0 to 5, to what extent do you (61 per cent) knew and had experienced the previ- agree this is a place where you’d stop and ous layout of Exhibition Road, and 39 per cent of socialise? [A subjective slider from 0 (com- the interviewed people lived or worked in the area, pletely disagree) to 5 (completely agree) was thus being considered regular users, while the available for the respondents]. remaining 61 per cent did not. Regarding demo- (5) When using the street, beyond which point graphics, the gender distribution was 43 per cent would you say cars have priority over pedes- female and 57 per cent male; 6 per cent were under trians? (a) Kerb, (b) drainage gully, (c) trees, 20 years, 59 per cent were between 20 and 39 years, (d) benches, (e) parking , (f) lamp posts, 24 per cent were between 40 and 65 years, and (g) carriageway, (h) none (cars don’t have 11 per cent were over 65 years of age. priority over pedestrians). (6) If crossing this part of the street at any point, would you say: (a) pedestrians have priority Case Study Selection over cars, (b) cars have priority over pedestri- ans, (c) they have equal priority, (d) I don’t In the last decade, a good number of streets have know. been redesigned by adopting shared space schemes (7) Did you know Exhibition Road before its in over 20 towns and cities across the UK, including refurbishment? (a) Yes, (b) no; (if yes) would Ashford, Bath, Bolton, , , Leeds, you prefer Exhibition Road in its previous London, Manchester, Newbury, Oxford, Plymouth, conventional layout? (a) Yes, (b) no. Poynton, Preston, Sheffield, Southampton and others. Some of these have been chosen as case Interviewers collected the answers on a tablet, studies in different investigations on shared space which was offered to the respondents to interact and similar street schemes that allow pedestrians freely through questions 2–5. and vehicles to share the same surface, namely those of Biddulph (2010, 2012b), Curl et al (2015), Ham- Data analysis and description of the sample mond and Musselwhite (2013), Havik et al (2015), Kaparias et al (2013, 2015), Moody and Melia (2014) Data from on-street interviews were analysed using and MVA Consultancy (2010, 2011). the Statistical Package for Social Sciences for further Exhibition Road, however, is the most well interpretation. Besides descriptive analysis, inferen- disseminated, popular and recognized among all tial statistical analysis was carried out using non- such cases. This was a very important reason for parametric methods: Mann–Whitney or Kruskal– selecting this case study, since it makes this work Wallis tests were used depending on whether more accessible and understandable to the urban

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Shared space streets design community, given that a large number of readers might have first-hand experience of the street. Besides, the project enjoys a good reputa- tion in the profession, having received both a Royal Institute of British Architects (RIBA) award and the European Prize for Urban Public Space Special Mention in 2012, and the Civic Trust Award in 2013 (Ruiz-Apila´nez and Arnaiz, 2013). The existence of previous studies on Exhibition Road (Kaparias et al, 2013, 2015; MVA Consul- tancy, 2012, 2013a, b;SYSTRA,2014)wasalso important for its selection, since this allows broader understanding of the case, with different studies complementing each other. Furthermore, Exhibi- tion Road receives intensive and heterogeneous pedestrian use, including visitors and locals, mak- ing it appropriate to investigate whether frequency of use and familiarity with shared space schemes influence street users. Lastly, Exhibition Road makes an interesting case study due to the number of similarities and differences that can be found along the full length of the street, including layout, adjacent land use, and traffic flow. All these specifications are detailed in the following section.

Description of the Case Study and Study Areas

Figure 1: Exhibition Road by the V&A and Science museum: Exhibition Road is located at the heart of a well- (above) before, and (below) after. known cultural centre in South Kensington, Lon- Source: Royal Borough of Kensington and Chelsea. don, providing access to some of the city’s most popular museums, namely the Victoria and Albert, Science, and Natural History Museums, The project was implemented from 2008 to 2011, as well as prestigious academic institutions such including a single kerb-free surface with no bar- as Imperial College London and the Royal Geo- riers and minimum street clutter, new street graphical Society. The street is about 800 m in lighting, new benches and trees, a 20 mph speed length, with its northern end reaching the edge of limit, two 4-m-wide pedestrian safe zones, one on Hyde Park and the southern end close to South each side of the street, delimited by visual and Kensington underground station. tactile lines which incorporate the gullies, and a The previous conventional dual-carriageway wide and direct at the inter- layout was extremely congested and showed high section with the heavily traffic-loaded Cromwell pedestrian and vehicular traffic with a streetscape Road. As a result, the street now has a fairly dominated by motor vehicles. The 24-m-wide consistent layout, although showing some differ- cross-section used to have two 4-m-wide footways ences in different sections regarding adjacent and a 16-m-wide , accommodat- building frontages and land use, as well as the ing three for street parking: one on each side street design itself, especially in relation to ele- and a central one, separating an oversized traffic ments of shared space, as broadly described in the lane on each side (Figure 1). RBKC promoted the next section. redesign of this street layout to improve the Using a set of similarities and differences, four former situation, which was considered ‘‘confus- different SAs were identified within Exhibition ing for visitors and unfriendly to pedestrians’’ Road, in addition to two other SAs with more (RBKC, 2012), and to recognize other needs of the conventional layouts from nearby Queen’s Gate place beyond those related to vehicular traffic. and Cromwell Road, considered as the control

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Ruiz-Apila´nez et al

activities. The location of the six SAs is shown in Figure 2, and they are further described below. The four SAs in Exhibition Road (SA1–4) show a 24-m-wide kerb-free, homogeneously paved sur- face, including a 4-m-wide pedestrian safe zone on each side of the street. Each SA also shows some specific characteristics, not only regarding adjacent land use and building frontages, but also in relation to the specific design of the street and the presence/absence of elements of shared space. In fact, they obtain different scores when using the shared space rating (SSR) system applied in the operational assessment report for the Department for Transport. This shared space classification questionnaire allows streets to be rated according to characteristics envisaged to encourage sharing. This method gives different numbers of points based on the absence of kerbs or other physical demarcations between pedestrian and vehicular areas, the homogeneity in surface colour treat- ment, the presence of other ‘‘public space charac- teristics’’, i.e. cafes/markets, benches, greenery or art, the non-demarcation of crossing points, and the absence of road markings, traffic lights, bollards, guard rails and street lamps (see MVA Consultancy, 2010, p. 2.4 for further information). The remaining two areas (SA5–6), however, show conventional layouts and therefore a significant physical difference from the previous four, but certain similarities in terms of land use, building frontages, vegetation and absence of benches. Figure 2: Location of the SAs in London (above) and South The average hourly vehicular traffic from 12:00 to Kensington (below): (1) Exhibition Rd–South, (2) Exhibition Rd– 14:00 for the six SAs was obtained from the above- Museum, (3) Exhibition Rd–University, (4) Exhibition Rd–North, described systematic counting procedure, whereas (5) Queen’s Gate–University; and (6) Cromwell Rd–Museum. the 85th percentile speed for the Exhibition Road Source: base maps obtained from ª Crown Copyright and cases only was extracted from phase 4 of the Database Right (2013). Ordnance Survey (Digimap Licence). Exhibition Road Monitoring Report (SYSTRA, 2014). group. The former control street is parallel to Exhibition Road with well-developed trees and Study Area 1 (SA1): Exhibition Road – South similar road width and traffic flow, similar land use along its length, and a more pedestrian-friendly The first SA (Figure 3) corresponds to the south layout than that offered by Exhibition Road before end of Exhibition Road, near the underground the change. Similar to Exhibition Road, Queen’s station. There is access-only one-way traffic, and Gate gives access to Imperial College London, but parking is not permitted except for loading and there are no museum entrances, representing a key unloading, so motor traffic volume is as low as difference in terms of pedestrian activity. This is 79 veh/h with 85th percentile speed of 14.1 mph. why the other SA in Cromwell Road was identified The south end connects with a perpendicular precisely at the main entrance of the Victoria and pedestrian street towards the underground sta- Albert Museum. Despite the higher traffic flow, this tion, and the north pedestrian access has one SA has a generous footway with plenty of sitting traffic-light crossing on each side of the street. facilities and some fully grown trees, making it Although there are no benches, some secondary potentially attractive to support stationary human seating is provided by the generous step around

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Shared space streets

Figure 3: SA1: Exhibition Road–South. The view from the Figure 4: SA2: Exhibition Road–Museum. The view near the northern part looking southwards. with Cromwell Road looking northwards. the underground air shafts. This sector is com- pletely surrounded by some small shops and lots of cafes and eateries on both sides of the street, providing plenty of tables and chairs to sit outside. There are some new trees planted on the west side of the street, and some medium-sized street lamps are aligned with the edges of the pedestrian safe zones. The shared space score for this sector is 31, as obtained from the SSR system.

Study Area 2 (SA2): Exhibition Road – Museums

The second SA (Figure 4) corresponds to the area between the Exhibition Road entrances to the Figure 5: SA3: Exhibition Road–University. The view near Victoria and Albert and Science Museums. A two- Imperial College Road looking northwards. way traffic area carrying medium vehicular vol- umes (573 veh/h) at speeds below 20 mph (18.8 mph) occupies the central east side of the street Road. The motor traffic volume (639 veh/h) is between the corresponding pedestrian safe zone slightly greater than in the preceding sector, SA2, and the high lampposts that are aligned with the but the speed is above 20 mph (22.8 mph) despite street’s central line. The delimitation of the safe the current 20 mph limit. The layout is almost zone on the east side of the street is reinforced by identical to that in SA2, but here there is no aligned trees and bollards situated at its edge. The vertical delimitation between the two-way traffic central part on the western side of the street is area and the pedestrian safe zone on the east side either occupied by perpendicular parking lots, of the street. There are no bollards and almost no bike racks and benches, or is left free. This sector of trees planted at the limit of the safe zone. The the street obtains a shared space score of 25, buildings on the east side are mainly houses and according to the SSR system. offices. It gets a shared space score of 30, according to the SSR system.

Study Area 3 (SA3): Exhibition Road – University Study Area 4 (SA4): Exhibition Road – North

The third SA (Figure 5) corresponds to the area in The fourth SA (Figure 6) corresponds to the front of Imperial College London on Exhibition northern part of Exhibition Road, to the south

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Ruiz-Apila´nez et al

Figure 6: SA4: Exhibition Road–North. The view near the Figure 7: SA5: Queen’s Gate–University. The view near the looking northwards. intersection with Imperial College Road looking northwards. of Hyde Park. It carries more traffic (998 veh/ h) than the other sectors of the street, and the speed (22.1 mph) is similar to that in SA3. Quite differently from the previous sectors of the street, the cross-section of the street is symmetrical. As happens along the whole street, except for the southern part, high lamp- posts are located along the central axis, but here they separate the two traffic streams that occupy the whole central part of the street. In addition, parallel parking is permitted on both sides of the street, by the pedestrian safe zones, which have trees planted along their edges. The buildings on both sides of the street are mainly residential, but there are some institu- Figure 8: SA6: Cromwell Road-Museum. The view from the tionalbuildingsaswell.Thesharedspacescore eastern part looking westwards. forthissectoris29,accordingtothe SSR system. Study Area 6 (SA6): Cromwell Road – Museum

Study Area 5 (SA5): Queen’s Gate – University The last SA (Figure 8) corresponds to the area in front of the main entrance to the Victoria and The fifth SA (Figure 7) corresponds to the area in Albert Museum at Thurloe Place, connecting front of Imperial College London on Queen’s Gate. Cromwell Road and Brompton Road, half a block Similarly to case study 3, one side of the street has away from Exhibition Road. The oversized car- university facilities and the other has buildings that riageway includes two traffic lanes in each direc- are mainly residential but include some offices with tion with stopping spaces for public transport and other professional services within. The layout is tourist buses on both sides and a central lane almost identical to that of Exhibition Road before the dedicated to left turns and taxis. Vehicular traffic change, but motor traffic, although still moderate, is volumes are significantly higher (2256 veh/h) significantly higher (1104 veh/h). The 30-m-wide than in any of the other cases, as is the speed cross-section is mainly symmetrical, with 4-m-wide based on mere observations. The width of the footways with trees and a 1-m-wide central median footway in front of the museum along this SA is where street lamps are located. Both carriageways 15 m. Besides three benches, the over 50-m-long include parallel parking on both sides, i.e. four lines entrance steps provide plenty of secondary seat- of parking in total, and an oversized traffic lane. ing. Mature and significantly sized plane trees are

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Shared space streets aligned by the carriageway all along the road but not in front of the steps.

Safety and Comfort

Supporters of shared space, such as Hamilton- Baillie (2008a), advocate such schemes for their good performance in terms of safety, based on the reported decrease in the number of incidents, accidents and injured people (Commission for Architecture and the Built Environment, 2007; Euser, 2006; Karndacharuk et al, 2014b). In relation to the redesigned Exhibition Road, the final RBKC report (SYSTRA, 2014) states no Figure 9: ‘‘In a scale from 0 to 5, how safe do you feel using this accidents and one or two daily incidents of street in relation to traffic?’’ Boxplot of the scores from 0 (not motorised vehicles stopping abruptly, based on safe at all) to 5 (very safe) given by pedestrians at each SA: (1) 5-h observation periods at six sections of the street Exhibition Rd–South, (2) Exhibition Rd–Museum, (3) Exhibition (four days in total, one for each of the four phases Rd–University, (4) Exhibition Rd–North, (5) Queen’s Gate– University, and (6) Cromwell Rd–Museum. of the study in May and November 2012 and 2013). There are no available data on the previous state, but it is claimed that the new layout is not unsafe, based on the small number of incidents and accidents reported. Shared space opponents have argued that the lack of incidents is due to people’s avoidance of the street or their use of it with greater caution, causing them stress and lack of comfort. To shed light on this discussion, as mentioned by Ham- mond and Musselwhite (2013), Kaparias et al (2012) and Moody and Melia (2014), pedestrians were asked to rate their perceived degree of safety and comfort in relation to traffic while using the street, in the six SAs. With regards to safety, the scores given by pedestrians are shown in Figure 9. The scores are consistently high, and it is remark- Figure 10: ‘‘In a scale from 0 to 5, how comfortable do you feel able how the median is almost identical for all six using this street in relation to traffic?’’ Boxplot of the scores SAs, with a value of almost 4. The similarity in the from 0 (not comfortable at all) to 5 (very comfortable) given by pedestrians at each SA: (1) Exhibition Rd–South, (2) Exhibition distribution of quartiles among all cases (except Rd–Museum, (3) Exhibition Rd-University, (4) Exhibition Rd– the conventional layout in SA5) is noticeable. In North, (5) Queen’s Gate–University, and (6) Cromwell Rd– these five cases, the top 25 per cent of scores is Museum. higher than 4.7. Except for SA1, less than 25 per cent of users at each location rated the perceived safety below 2.5. significant differences were detected between the Regarding comfort, the scores given by pedestri- SAs (Kruskal–Wallis test: v2 = 2.698, p [ 0.05 and ans showed values consistently higher than for v2 = 2.706, p [ 0.05, respectively). According to the safety, with medians between 4.3 and 5 for all cases, user survey, and despite the significant differences and it is noteworthy that they reach that maximum in design, vehicular volume and speed, the different value at two locations, SA4 and SA6. At each street layouts were perceived as safe and comfort- location, less than 25 per cent of people ranked the able—indeed very safe and comfortable—at least perceived comfort below 3.4 (Figure 10). when carrying the observed traffic flows. Further statistical analysis confirmed the remark- Looking at the amount of people walking able similarities found in the box-plot graphs for through the different sectors during the lunchtime both perceived safety and comfort, since no period, SA1 and SA2 were the busiest,

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Ruiz-Apila´nez et al accommodating around 2000 pedestrians/h (2290 and 1938 ped/h, respectively), whereas a lower number of people walked through sectors SA3 and SA6 (1347 and 1038 ped/h, respectively), and an even lower number through SA5 and SA4 (660 and 423 ped/h, respectively). Many different factors influence the pedestrian flow for a particular street, but looking at these numbers and understanding the location of the different attractors in the area, it is difficult to argue that the new implementation of Exhibition Road is making people avoid the street, as has been suggested for other shared space areas, such as the one in Ashford, Kent (Moody and Melia, 2014).

Influence of familiarity with shared space

Regarding the attitudes, perceptions and concerns of pedestrians regarding shared space, Hammond and Musselwhite (2013) suggest that pedestrians’ confidence in relation to motor traffic might be linked to their familiarity with the non-conventional layout. Kaparias et al (2015, p. 125) refer to a so- called ‘‘settling-down period’’, after which users might become accustomed to the new design. Intuitively, it does sound reasonable that people might need a certain period to get used to the new layout, and once they are familiar with the new design, they become more confident using it. How- ever, the good performance of shared space with regards to traffic safety is based on the uncertain degree of awareness of the users in dealing with the intentionally less clear rules in the new layout, which might be thought to decrease once they become more familiar with it. Consequently, it is pertinent to investigate whether regular and non- Figure 11: Boxplots of the scores on safety (above) and comfort regular users (visitors) have a different perception of (below) from 0 (not safe/comfortable at all) to 5 (very the safety and comfort provided by different layouts safe/comfortable) given by visitors and regular users at each and, if so, to what extent. SA: (1) Exhibition Rd–South, (2) Exhibition Rd–Museum, (3) The scores on safety and comfort given at each Exhibition Rd-University, (4) Exhibition Rd–North, (5) Queen’s of the SAs, differentiating between visitors and Gate–University, and (6) Cromwell Rd–Museum. regular users, are shown in Figure 11. In terms of safety, SA2 and SA3 on Exhibition Road were rated slightly higher by regular users than -0.383, p [ 0.05). In terms of comfort, regular by visitors, but differences were not statistically users consistently provided lower scores for all significant (Mann–Whitney test: z =-0.844, p [ SAs, being of statistical significance only for SA1 0.05 and z =-0.104, p [ 0.05, respectively). Con- (z =-3.392, p = 0.01) but not any other SA versely, the higher scores given by non-regular (p [ 0.05 in all cases, z =-0.186, -1.41, -0.118, users at the other four locations were statistically -1.021, and -0.632 for SA2–6). significant for SA1, SA4 and SA5 (z =-3.621, Focussing on the four occasions where scores p \ 0.001; z =-3.161, p = 0.002; and z =-2.752, were significantly different, and taking into p = 0.006, respectively), but not for SA6 (z = account the different layouts and performance of

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Shared space streets the three locations where this occurred, the results of this study do not support the notion that shared space users gradually and over time become more confident in using this type of street layout. What the analysis of the surveys implies though is that regular users are more critical about the safety and comfort performance of the streets. It is therefore suggested that people examine more critically streets with which they are familiar, which is something that should be taken into consideration in future research.

Pedestrians’ Perceived Realm and Assumed Priority

Figure 12: ‘‘When using the street, beyond which point would According to the original concept of shared space, you say cars have priority over pedestrians?’’ Percentages of in such layouts, it is desired that traffic rules will answers obtained at each SA: (1) Exhibition Rd–South, (2) become uncertain to users to some extent, so they Exhibition Rd–Museum, (3) Exhibition Rd-University, (4) have to negotiate their movements, encouraging Exhibition Rd–North, (5) Queen’s Gate–University, and (6) use of the whole width of the street. It is crucial in Cromwell Rd–Museum. this regard to recognise how different layouts in reality inform pedestrians, and how they perceive street function in relation to motor traffic. 69 per cent of respondents located the limit of the Two of the questions included in the question- pedestrian domain at the kerb while 22 per cent naire were aimed at investigating how pedestrians opted for the parking area. understand the functional distribution of the In the shared space layouts, the answers varied street’s cross-section and the existing hierarchy a lot between the SAs. SA4, which despite having between them and drivers. The answers given by a levelled cross-section also has linear parking the pedestrians are discussed and put in context in between the designated safe zones and the traffic the two corresponding parts of this section. thoroughfare, is mostly understood as a conven- tional layout. Similarly to SA5, 62 per cent of respondents in SA4 set the limit of their domain at Limits of pedestrian domain the drainage gullies (similar to the 69 per cent who opted for the kerb in SA5) and 20 per cent at the Firstly, users were asked to indicate beyond which parking area (very similar to the 22 per cent for point they felt cars had priority over pedestrians, SA5). The perceived limit changed drastically in choosing between ‘‘none: cars don’t have priority SA3, where only 18 per cent referred to the gully over pedestrians’’ and a set of physical elements while more people (31 per cent) mentioned the (lamp posts, kerb, drainage gully, benches, trees) parking area (perpendicular and far less continu- and conventional function-related areas (parking ous here) and 54 per cent set the limit in the area, carriageway). The distribution of answers at middle of the road (lamp posts and carriageway different locations is plotted in Figure 12. Given refer to the same limit here). the asymmetrical layout of the sections and to This shift in user perception is evident for SA2, better understand the answers, note that all where 70 per cent of respondents referred to the interviews for Exhibition Road were carried out middle of the street as the boundary of the on its western side. pedestrian domain, and 18 per cent stated that Examining the two conventional layouts, for cars do not have priority over pedestrians at any SA6 it is shown that over 90 per cent of pedestri- point. For SA1, the same 88 per cent of pedestrians ans set the limit of the pedestrian domain at the chose these two options, only 35 per cent thought kerb or the carriageway, which correspond to the that vehicles did not have priority at all, and same spatial limit in this specific conventional 53 per cent referred to the carriageway. This layout with no parking. For SA5, where parking is progressive extension of the perceived pedestrian permitted by the footways, it is noticeable that limit is confirmed by the statistical analysis,

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Ruiz-Apila´nez et al showing a significant direct correlation with the Positioning in the street and assumed priority different locations (Kendall’s tau-c test: sc = 0.631, p \ 0.001). It is inferred that a broader pedestrian To clarify further the above-mentioned issue, domain is perceived in shared space streets than in pedestrians were also asked more precisely and those with conventional layouts. directly about their perceived crossing priority: If Only in SA1 and SA2 did a significant number they wanted to get to the opposite side of the street of people (35 and 18 per cent, respectively) state at any given point along each of the SAs, who had that cars do not have priority over pedestrians at priority: pedestrians or drivers? People could any point within the street section. Considering choose to say that they did not know, state that how shared space is conceived to work, it is neither of them had priority over the other, or evidentthatitisnotnecessarilyunderstoodin select pedestrians or drivers. this way by the large majority of pedestrians. Only 8 out of the total 305 respondents Similarly to other authors, Karndacharuk et al answered ‘‘I don’t know’’ (2.6 per cent). Looking (2014a) argue that, over time, pedestrians might at each individual SA, the maximum number of become more assertive and take greater control of pedestrians who reported to feel unclear about the street space, but based on this study, there is the priority was found for SA2, although there no evidence to support such an argument. As were only three (5 per cent). This low number of shown in the previous section, time itself does not people who were not sure about the priority is in provetoworkinthedirectionthatonemight conflict with the behavioural assumption that hope things to change. users might get confused in a shared space As described in the ‘Methodology’ section, environment. Quite the contrary, this research while counting pedestrians to determine the flows, shows that people seem to be confident about a those walking along the carriageway were modal crossing hierarchy when they use the counted separately. The different percentages street. The answers obtained at each location are observed along the different SAs verify some of plotted in Figure 13. the discussion on the results of the survey. In the two conventional layouts, SA5 and SA6, all pedestrians walking along the street remained on the footways, and in SA4, pedestrians walked predominantly along the designated safe zones. Similarly, in SA3, people used the whole section except for the carriageway. However, in SA1 and SA2, a noticeable percentage of pedestrians (7 and 12 per cent, respectively) walked along the car- riageway, which in these sectors corresponds with the part between the centre of the road and the safe zone along the east-side buildings. Although both SA1 and SA2 did carry a large number of people, the highest flow observed (SA2) was no higher than 7 people/minÁm, considering only the width of the safe zones, not the whole width of the street. This flow is much smaller than those observed to make people walk out of the footways and use the carriageway in conventional streets, around 13 people/minÁm, according to Jacobs (1993). It is therefore argued that, if people walk down the carriageway along this sections, it Figure 13: ‘‘If crossing this part of the street at any point, would is not because there is a lack of safe or comfort- you say: (a) pedestrians have priority over cars, (b) cars have able pedestrian space available. It is suggested that priority over pedestrians, (c) they have equal priority, or (d) I these people feel they have the right to use this don’t know’’ [option (d) not plotted due to lack of responses] Percentages of answers obtained at each SA: (1) Exhibition Rd– part of the road at these two locations, yet they South, (2) Exhibition Rd–Museum, (3) Exhibition Rd- step out and occupy a different part of the street University, (4) Exhibition Rd–North, (5) Queen’s Gate– when a motor vehicle approaches. University, and (6) Cromwell Rd–Museum.

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Shared space streets

To understand how pedestrians’ assumptions significant assessment, the comparisons focus on relate to the applicable traffic regulations, it is equal-length stretches that include designated important to point out that, in the two conven- pedestrian crossings, except for SA3, which has tional streets, where it might be assumed that all no designated crossing along the whole stretch. users are aware that cars do have legal priority, The registered 5-min interval of maximum less than 75 per cent agreed with this assumption. crossing activity at each location is presented in This suggests that about 20–30 per cent of people Figure 14, indicating the number of those using the might be answering this question based on their designated pedestrian crossings and those cross- own behaviour or personal values regarding ing at other points of the street. SA4 is not taken transportation modal priorities, rather than con- into consideration since its low pedestrian activity, sidering actual traffic regulations. especially in terms of crossing, makes it irrelevant Considering the people who believed cars had for the study. priority over pedestrians, three groups can be Analysis of the traces and the number of differentiated among the SAs. In the first group crossings reveals two facts. First, the highest (SA4–6), over 70 per cent of pedestrians considered absolute and relative numbers of crossings outside that cars had priority. In the second group (SA2 the designated crossings were found for SA1 and and SA3), approximately 50 per cent of pedestrians SA2, being much higher than for SA5 and SA6. agreed with vehicular predominance, and in the Second, the traces are much more homogeneously third group (SA1), only 30 per cent stated that cars distributed along SA1–3 than along SA5 and SA6. had priority over pedestrians. Again, there are The observed crossing patterns suggest that large discrepancies across the four shared space pedestrians behave more freely and make sections studied. broader use of the street surface in shared space To contrast what pedestrians reported with their streets than in those with conventional layouts. In actual behaviour, crossing behaviour was also this regard, it is informative to compare SA3 observed in the six areas. To establish a more and SA5 to find out how differently a similar

Figure 14: Pedestrians crossing the street during the busiest registered 5-minute interval at each SA: (1) Exhibition Rd–South, (2) Exhibition Rd–Museum, (3) Exhibition Rd-University, (4) Exhibition Rd–North, (5) Queen’s Gate–University, and (6) Cromwell Rd– Museum.

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Ruiz-Apila´nez et al number of people, entering and leaving the Figure 16: Maximum registered stationary activity [peoplec university campus, behave on the equivalent standing (orange), sitting-primary (light blue), sitting-secondary street stretches to the west and east of Imperial (dark blue), and sitting-tertiary (deep purple)] at each SA: (1) Exhibition Rd–South, (2) Exhibition Rd–Museum, (3) Exhibition College London, on Exhibition Road and Queen’s Rd-University, (4) Exhibition Rd–North, (5) Queen’s Gate– Gate, respectively. University, and (6) Cromwell Rd–Museum. The great contrast observed among the differ- ent SAs and within those on Exhibition Road confirms that the actual street designs do make a cent reported ‘‘no’’ to the same question asked by difference in how the space is used and also Moody and Melia (2014) for another case in shows how the different treatments of shared Ashford, UK. space do impact differently on pedestrians’ per- Due to the significant disagreement between the ception and behaviour. findings of various studies and to allow compar- ison with them, the questionnaire designed for this investigation included the question ‘‘In a scale Socialisation from 0 to 5, to what extent do you agree this is a place where you’d stop and socialise?’’ The scores Shared space is claimed to foster social activity people provided at each location are shown in (Department for Transport, 2011; Hamilton-Baillie, Figure 15. 2008a), although available research is not conclu- In this case, the results vary again greatly across sive on this matter. On the one hand, Biddulph SAs in general and also across the shared space (2012a, b) found that shared streets in residential ones. SA1 is seen by the majority of pedestrians as areas (UK home zones) were more successful than a great place to socialise, with over 50 per cent of conventional traffic-calming solutions, although people giving the maximum score. In both SA2 this would mainly benefit children and those and SA3, the median was slightly over 3 and the looking after them, and Karndacharuk et al range of responses varied greatly, from 0 to 5. (2013) found increases in pedestrian occupancy This variation was also found for the remaining after shared space street implementation. On the three locations (SA4–6), although in these cases other hand, Curl et al (2015) did not find evidence the scores were lower and their medians remained of positive change in outdoor social activity. While between 1.1 and 1.9. Further statistical analysis Hammond and Musselwhite (2013) found that confirmed the described significant differences 57 per cent of users of a shared space scheme in between these three different groups: SA1, SA2–3 2 Hereford, UK would stop and socialise, 65 per and SA4–6 (Kruskal–Wallis test: v = 60.796, p \ 0.001), while revealing no significant differ- ences in intention to socialise between regular and non-regular users at each location (Mann–Whit- ney test: p [ 0.05 in all cases, z =-1.244, -1.796, -0.463, -0.325, -0.9 and -0.606, for SA1–6, respectively). In this sense, ground-floor uses and frontages greatly affect the perception of pedestrians (Borst et al, 2008;Gehl,2010;Jacobs,1993). Therefore, the perceived differences between areas with similar ground-floor use, e.g. SA2 and SA6 (both sharing museum entrances) or SA3 and SA4 (both sharing university campus entrances), suggest that these significant discrepancies in users’ perceptions are related to the substantial changes in street design, and therefore that shared space Figure 15: ‘‘In a scale from 0 to 5, to what extent do you agree schemes have potential to modify the social this is a place where you’d stop and socialise?’’ Boxplot of the scores from 0 (completely disagree) to 5 (completely agree) appeal of the street. given by pedestrians at each SA: (1) Exhibition Rd–South, (2) The actual social activity at the different SAs Exhibition Rd–Museum, (3) Exhibition Rd–University, (4) was systematically mapped to contrast with peo- Exhibition Rd–North, (5) Queen’s Gate–University, and (6) ple’s perceptions. The observations of how people Cromwell Rd–Museum. occupied the street while engaging in more

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Shared space streets

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Ruiz-Apila´nez et al optional and social stationary activities partially street participated in the survey, so it might be confirm the discussion of the survey results and possible that those not so in favour of the new layout provide further information for analysis. To facil- could be avoiding the street and, thus, had not been itate comparison between locations, the observed taken into consideration. stationary human activities, including standing It can also be argued that people could be more and sitting (using primary, secondary or tertiary likely to be in favour of the implemented scheme, seating), are plotted for equal-length stretches of just because of its newness, in opposition to the the six SAs, showing the busiest snapshot from poorer state that the previous, old layout might 12:00 to 14:00 (Figure 16). have presented. Probably in order to cancel this These snapshots emphasise the difference new–old dichotomy, Moody and Melia (2014, between the complete lack of activity in SA4 and p. 7) reformulated the question and asked people SA5 and the bustle in SA1, with lots of people if they would ‘‘prefer traditional pavements and sitting outdoors, despite the lack of primary crossings’’ instead. In this case, 64 per seating. This of course has to do with the existing cent of people preferred the more conventional ground-floor uses of the buildings, such as cafes layout. and restaurants, which provide plenty of tertiary The present study considers all those who seating. However, the use of the shared space knew Exhibition Road before the new implemen- available beyond the safe zones (former footways) tation, when using either the shared space suggests a symbiotic relationship between the schemes (SA1–SA4) or conventional layouts adjacent land uses and public space uses, with a (SA5 and SA6). Pedestrians were asked if they strengthening, intensifying effect. preferred the previous layout of the street. A Observations however are not helpful to con- substantial majority of 72 per cent preferred the firm the differences observed by users between new shared space layout. This percentage turned SA2 and SA3 (both sociability medians above 3.0), out to be remarkably consistent across all loca- and SA6 (below 2.0). It is suggested that these two tions: 71 per cent in the case of those using the areas of Exhibition Road (SA2 and SA3) do have shared space and 73 per cent of those walking in some undeveloped potential to engage people in the conventional streets. stationary activities. It can easily be imagined that Considering these results and some of the incrementing seating possibilities and introducing differences found in users’ perceptions and the other attractors, e.g. street food vendors (Whyte, performance of the different locations, it would 1980), would probably result in a decisive trans- have been informative to ask whether there was a formation of these locations in terms of social section or layout that users particularly preferred activity. over the others. This will be taken into consider- ation in future research.

Users’ Overall Outlook Conclusions As highlighted above, shared space is to some extent a rather controversial approach to street This study aimed to further enhance our under- design, about which people have expressed differ- standing of how safe and comfortable pedestrians ent opinions regarding its key aspects, sometimes feel in a shared space environment, as well as our varying greatly depending on the particularities of understanding of the potential of this street design each design and location, as also shown by this approach for enhancing public life. The results study. To grasp the final overall opinion of the show that average pedestrians find the shared general public, and in the absence of data on the space layouts of the analysed SAs both safe and prior-to-implementation state, Hammond and comfortable in relation to traffic, at least as safe Musselwhite (2013) asked users to self-report a and comfortable as more conventional streets. before and after comparison. Seventy-four per cent However, and rather differently from what has of those who had experienced the street before and been suggested by other authors previously, this after preferred the new, shared space design. This research indicates that the perception and attitude clear positive attitude towards the new layout was of pedestrians do not improve with time nor taken with some reservation by the same research- depend on their familiarity with a specific shared ers, arguing that only people actually using the space scheme.

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Shared space streets

This does not mean that people are non-respon- environment and surroundings is necessary to sive or unaffected by street design. This research assess any future implementations. reveals that, at least for these six SAs, the very After over a decade of shared space practice, opposite is true by, firstly, showing how the including a good number of successful examples perceived limit of the pedestrian domain varies and some research, it is suggested that more greatly depending on the street layout, and research is required, probably not to keep ques- secondly, by illustrating the movements along tioning the core concept itself, but rather to test and across the streets and how the perceived and evaluate the different outcomes, to better crossing priority between pedestrians and drivers understand the particularities of different cases in differs among the SAs. terms of the design considerations, user percep- The capacity of shared space schemes to tions and performance. enhance public life has been investigated by In this regard, the lack of consistent and com- previous research, but the outcomes have not parable before–after case studies as well as the been conclusive. This study confirms that street importance of enabling accurate comparison design is a powerful tool to transform the built between future studies should be emphasised. environment and influence people’s perception The present investigation has made an effort to and use of the street space in relation to stationary establish a dialogue with other researchers by public and social activity. However, ground-floor carefully considering their methods and findings building frontages and uses also play a decisive to incorporate or discuss them as appropriate. The role in this regard, being even more decisive spirit has been to make this work as useful as sometimes. possible for the urban design community, and we Considering this research and that by other recommend that forthcoming research should authors cited throughout the text, it is suggested operate in a similar way. that shared space can no longer be approached as Lastly, it must be stressed that the shared space a homogeneous element and that it is necessary to approach to public space design encourages a understand that questions such as ‘‘Do shared ‘‘focus shift from project to process’’ (Besley, 2010, space streets work?’’, ‘‘Are shared space streets p. 6) and highlights the importance of including safe?’’, and the like, may no longer be valid, in the participatory design techniques. However, this same way that questions such as ‘‘Do pedestrian research and most other research published to streets work?’’ or ‘‘Are highways safe?’’ would be date focusses on evaluation of implemented pro- considered too general or unspecific to be jects and not the implementation process. This answered in any accurate way. should be taken into consideration not only by In the view of recent efforts to create simplistic public administrators and designers but also for evaluation tools to quantify the degree of shared future research. space, such as that used in the report for the Department for Transport (MVA Consultancy, 2010), this research suggests that shared space Acknowledgements should rather be considered as a street and public space design approach that lies beyond the mere This work was supported by the University of inclusion/exclusion of a number of features, e.g. Castilla-La Mancha under the CYTEMA-PUENTE traffic signals, levelled surface, kerbs, clutter and grants. The authors would like to acknowledge the demarcations, benches, greenery or art, and traffic invaluable help with the street audits provided by signals. Conversely, more comprehensive methods Mayte Arnaiz, Lucy Donegan and Fanni Kos- including both qualitative and quantitative tech- torous, and would also like to thank the two niques are required, some of which have been anonymous reviewers for their detailed evalua- included in that same report. tion, constructive attitude and valuable comments. Based on some of the similarities and differences found along Exhibition Road, it must be empha- sized that users’ perceptions and the performance References of shared space cannot be advanced based on the accumulation of ‘‘elements of shared space’’ (Ka- Appleyard, D., Gerson, M.S. and Lintell, M. (1981) Livable Streets. Berkeley, CA: University of California Press. parias et al, 2013, p. 115). Rather, it is the street Besley, E. (2010) Kerb Your Enthusiasm. Why Shared Space Doesn’t layout as a whole that makes a difference. Deep Always Mean Shared Surface, and Other Stories. London: and detailed understanding of its design, Parliamentary Advisory Council for Transport Safety.

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International Ruiz-Apila´nez et al

Biddulph, M. (2010) Evaluating the English home zone initia- shared space. Transportation Research Part F: Traffic Psychol- tives. Journal of the American Planning Association 76(2): ogy and Behaviour 30: 115–127. 199–218. Karndacharuk, A., Wilson, D.J. and Dunn, R.C.M. (2013) Biddulph, M. (2012a) Radical streets: The impact of innovative Analysis of pedestrian performance in shared-space envi- street designs on liveability and activity in residential areas. ronments. Transportation Research Record 2393: 1–11. Urban Design International 17: 178–205. Karndacharuk, A., Wilson, D.J. and Dunn, R.C.M. (2014a) A Biddulph, M. (2012b) Street design and street use: Comparing review of the evolution of shared (street) space concepts in traffic calmed and home zone streets. Journal of Urban Design urban environments. Transport Reviews 34(2): 190–220. 17(2): 213–232. Karndacharuk, A., Wilson, D.J. and Dunn, R.C.M. (2014b) Borst, H.C., Miedema, H.M.E., de Vries, S.I., Graham, J.M.A. Safety performance study of shared pedestrian and vehicle and van Dongen, J.E.F. (2008) Relationships between street space in New Zealand. Transportation Research Record 2464: characteristics and perceived attractiveness for walking 1–10. reported by elderly people. Journal of Environmental Psychol- Methorst, R., Gerlach, J., Boenke, D. and Leven, J. (2007) Shared ogy 28(4): 353–361. space: Safe or dangerous? A contribution to objectification of Commission for Architecture and the Built Environment. (2007) a popular design philosophy. In: Proceedings of Walk21 This Way to Better Streets: 10 Case Studies on Improving Street Conference in Toronto, October, pp. 1–17. Design. London. Moody, S. and Melia, S. (2014) Shared space: Research, policy Curl, A., Ward Thompson, C. and Aspinall, P. (2015) The and problems. Institution of Civil Engineers (Transport) 167: effectiveness of ‘‘shared space’’ residential street interven- 384–392. tions on self-reported activity levels and quality of life for MVA Consultancy: Shore, F. and Uthayakumar, K. (2010) older people. Landscape and Urban Planning 139: 117–125. Designing the Future. Shared Space: Operational Assessment. Department for Transport. (2011) Shared Space—Local Transport London: Department for Transport. Note 1/11. London: TSO. MVA Consultancy: Dickens, L., Healy, E., Plews, C. and Euser, P. (2006) The Laweiplein. Evaluation of the Reconstruction Uthayakumar, K. (2011) Shared Space: Qualitative Research. into a Square with Roundabout. Leeuwarden: Noordelijke London: Department for Transport. Hogeschool. MVA Consultancy: Healy, E. and Fuller, A. (2012) Exhibition GDBA. (2006) Shared Surface Street Design Research Project. The Road Monitoring. Phase 1. London: Royal Borough of Kens- Issues: Report of Focus Groups. Reading: Guide Dogs for the ington and Chelsea. Blind Association. MVA Consultancy: Fuller, A. and Healy, E. (2013a) Exhibition Gehl, J. (2010) Cities for People. London: Island. Road Monitoring. Phase 2. London: Royal Borough of Kens- Hamilton-Baillie, B. (2004) Urban design: Why don’t we do it in ington and Chelsea. the road? Modifying traffic behavior through legible urban MVA Consultancy: Fuller, A. (2013b) Exhibition Road Monitoring. design. Journal of Urban Technology 11(1): 43–62. Phase 3. London: Royal Borough of Kensington and Chelsea. Hamilton-Baillie, B. (2008a) Towards shared space. Urban Parkin, J. and Smithies, N. (2012) Accounting for the needs of Design International 13(2): 130–138. blind and visually impaired people in public realm design. Hamilton-Baillie, B. (2008b) Shared space: Reconciling people, Journal of Urban Design 17: 135–149. places and traffic. Built Environment 34(2): 161–181. Ruiz-Apila´nez, B. and Arnaiz, M. (2013) The Exhibition Road Hamilton-Baillie, B. (2014) Espacio compartido. Paisea 29: 6–12. Project: un espacio compartido en el centro de Londres [The Hammond, V. and Musselwhite, C. (2013) The attitudes, Exhibition Road Project: A shared space in central London]. perceptions and concerns of pedestrians and vulnerable Ciudad y Territorio Estudios Territoriales XLV(178): 803–810. road users to shared space: A case study from the UK. Ruiz-Apila´nez, B., de Uren˜a, J.M. and Solı´s, E. (2014) La Journal of Urban Design 18(1): 78–97. revitalizacio´n de la calle: estrategias basadas en la remod- Hass-Klau, C. (1990) The Pedestrian and City Traffic. London: elacio´n [Revitalising the street: Redesign strategies]. Ciudad y Belhaven. Territorio Estudios Territoriales XLVI(181): 393–411. Havik, E.M., Steyvers, F.J., Kooijman, A.C. and Melis-Dankers, Shared Space. (2005) Shared Space: Room for Everyone. Leeuwar- B.J. (2015) Accessibility of shared space for visually impaired den: Interreg IIIB Project ‘Shared Space’. persons: A comparative field study. British Journal of Visual Shared Space. (2008a) Shared Space: From Project to Process. Impairment 33(2): 96–110. Leeuwarden: Interreg IIIB Project ‘Shared Space’. Imrie, R. (2012) Auto-disabilities: The case of shared space Shared Space. (2008b) Shared Space: Final Evaluation and Report. environments. Environment and Planning A 44(9): 2260–2277. Leeuwarden: Interreg IIIB Project ‘Shared Space’. Jacobs, A.B. (1993) Great Streets. Cambridge, MA: MIT Press. Shared Space, Hamilton-Baillie, B., Zingstra, A., Lont, D., Kaparias, I., Bell, M.G.H., Miri, A., Chan, C. and Mount, B. Schwertmann, W., Goedejohann, K., Pitchford, J., et al (2008) (2012) Analysing the perceptions of pedestrians and drivers Shared Space: Partner Publication. Leeuwarden: Interreg IIIB to shared space. Transportation Research Part F: Traffic Project ‘Shared Space’. Psychology and Behaviour 15(3): 297–310. SYSTRA: Kumar, J. and Healy, E. (2014) Exhibition Road Kaparias, I., Bell, M.G.H., Dong, W., Sastrawinata, A., Singh, Monitoring. Phase 4. London: Royal Borough of Kensington A., Wang, X. and Mount, B. (2013) Analysis of pedestrian- and Chelsea. vehicle traffic conflicts in street designs with elements of The Royal Borough of Kensington and Chelsea. (2012) The shared space. Transportation Research Record 2393: 21–30. Exhibition Road Project. https://www.rbkc.gov.uk/subsites/ Kaparias, I., Bell, M.G.H., Biagioli, T., Bellezza, L. and Mount, exhibitionroad.aspx, accesed 11 October 2013. B. (2015) Behavioural analysis of interactions between Whyte, W.H. (1980) The Social Life of Small Urban Spaces. New pedestrians and vehicles in street designs with elements of York: Project for Public Spaces, Inc.

ª 2017 Macmillan Publishers Ltd. 1357-5317 URBAN DESIGN International