Cross-cultural public space inclusivity: a multidimensional analysis of user satisfaction in Nottingham, Jeddah, and Lucknow

1 Introduction

Urban public spaces are central to fostering social interaction, civic engagement, and urban liveability (Gehl and Svarre, 2013; Mehaffy, 2021). Beyond their physical function, they operate as socio-spatial environments where cultural identities are expressed, and social relations are negotiated (Jacobs, 1992). However, the extent to which public spaces support inclusivity, accessibility, and a sense of publicness varies significantly across cultural and urban contexts. Although physical and morphological attributes such as accessibility, infrastructure, and spatial configuration have received considerable attention as determinants of public-space quality (Shafique, 2024), recent studies highlight the growing significance of social and perceptual dimensions, notably perceived publicness (defined operationally in Section 2.2), belonging, and cultural identity, in shaping user satisfaction (de León et al., 2020; Addas et al., 2021; Abid and Haque, 2024). Moreover, public space is increasingly understood as a socially constructed domain, where governance structures, cultural norms, and everyday practices mediate access and participation (Susewind, 2020; Thiollet and Assaf, 2021).

Despite these advances, important gaps remain. Much of the literature is based on single-context studies, limiting multi-cultural generalizability (Addas and Maghrabi, 2021; Wang et al., 2025). Additionally, there is limited comparative research that examines physical, social, and perceptual aspects of public space experience within a unified analytical framework (Addas, 2022; Salama et al., 2024). To address these gaps, this study adopts a comparative and user-centered approach to examine public space inclusivity and user satisfaction across three urban contexts: Old Market Square in Nottingham, Al Balad in Jeddah, and Hazratganj in Lucknow. By integrating accessibility, safety and comfort, inclusivity and equity, perceived publicness, and cultural identity, the study advances a holistic understanding of public space quality. The following section reviews the literature on integrated frameworks, publicness, accessibility, cultural identity, and multi-cultural urban design, providing the foundation for the research gap addressed in this study.

1.1 Research gap

Despite extensive research on urban public spaces, several critical gaps remain in existing literature. First, much of the research on public space adopts single-context or mono-cultural perspectives, often focusing on cities within the Global North or specific regional settings. This limits the ability to generalize findings or understand how public space experiences vary across diverse socio-cultural environments (Addas et al., 2021; Wang et al., 2025). Comparative cross-cultural studies remain relatively underdeveloped, particularly those employing a consistent methodological framework across multiple global contexts (Addas et al., 2021; Addas, 2022; Wang et al., 2025). Second, although recent studies advocate for interconnected and person-environment-behavior frameworks, many empirical investigations continue to prioritise physical and morphological attributes, such as accessibility, infrastructure, and spatial configuration, while giving insufficient attention to the relationships among spatial provision, user needs, behavior, and perception (de León et al., 2020; Addas et al., 2021; Shafique and Majid, 2026; Wang et al., 2025).

Concepts such as perceived publicness, belonging, and cultural identity are often discussed theoretically but are rarely operationalized and tested quantitatively within a unified analytical model (de León et al., 2020; Wang et al., 2025). Third, the relationship between objective spatial qualities and subjective user perceptions remains insufficiently understood. Existing research frequently assumes a direct link between improved design and enhanced user satisfaction; however, evidence suggests that this relationship is mediated by cultuemerging ral norms, governance structures, and social dynamics (Addas et al., 2021; Salama et al., 2024). There is a lack of unified studies that simultaneously examine design, governance, and perception-based variables in shaping public space outcomes (de León et al., 2020; Salama et al., 2024).

Fourth, while inclusivity is widely acknowledged as a central goal of public space design, there is limited consensus on how to measure inclusivity in a comparable and systematic manner across different contexts (Addas et al., 2021; Salama et al., 2024). Existing evaluation tools are often context-specific or lack the ability to capture the composite and culturally contingent nature of inclusivity. Finally, methodological limitations persist in the literature, with many studies relying on either qualitative case studies or isolated quantitative indicators, without integrating robust statistical validation techniques such as factor analysis and cross-site comparative modeling (Salama et al., 2024; Wang et al., 2025). This creates a gap in empirically validated, scalable frameworks capable of supporting both academic inquiry and practical urban design decision-making. Importantly, while the individual dimensions of accessibility, safety and comfort, inclusivity and equity, perceived publicness, and cultural identity are each well established in the public space literature, their simultaneous empirical operationalization and diverse cultural testing within a single validated quantitative model remains open for further empirical contribution. The research therefore examines how these dimensions interact within a single comparative framework. To address these identified gaps, this research adopts a cross-cultural, integrated, and user-centered approach that integrates spatial, social, and perceptual dimensions of public space quality.

1.2 Research contribution

This study makes three contributions. The first is theoretical: by comparing Nottingham, Jeddah, and Lucknow within one analytical framework, the study shows how the experience of public space differs across Western European, Gulf, and South Asian settings, and treats the five dimensions as interacting aspects of quality rather than separate ones.

The second is methodological. The study operationalises and tests all five dimensions together within a single validated cross-cultural measurement model. It also sets out the Cross-Cultural Public Space Inclusivity Index (CCPSII). The index does not introduce new constructs; it brings established accessibility, safety, inclusivity, publicness, and cultural-identity measures into one composite, transparent analytical tool for comparison across contexts.

The third is practical. The findings indicate that accessibility on its own does not secure a positive experience, and that safety, social interaction, and cultural relevance matter alongside it. This is relevant for planners and policymakers working across different cultural settings.

Research aim: This study aims to provide a cross-cultural analysis of user experiences in urban public spaces, showing similarities and differences in perceptions of inclusivity, equity, and publicness across diverse socio-cultural contexts, and to develop a holistic, user-centered framework for evaluating and enhancing inclusive public spaces.

1.2.1 Research questions

• To what extent do users' perceptions of accessibility, safety and comfort, inclusivity and equity, cultural identity, and overall sense of belonging differ across Old Market Square (Nottingham), Al Balad (Jeddah), and Hazratganj (Lucknow)?
• How do the five dimensions of public space experience interrelate, and which dimensions show the strongest associations across the three cultural contexts?
• How can the identified dimensions and cross-cultural patterns be synthesized into a holistic, user-centered evaluative framework for assessing inclusive public spaces, and what specific design, management, and governance implications arise for urban planners and designers in diverse contexts?

2 Literature review and theoretical background

2.1 Multidimensional frameworks for public space quality

A central debate in public space research is whether quality can be captured through single-variable assessments, such as visibility, aesthetics, or amenity provision, or whether it requires an integrated framework spanning physical, social, governance, and symbolic dimensions. The weight of evidence now favors the latter position: studies across heritage, governance, and cross-cultural contexts converge on the finding that no single variable adequately predicts user experience (Heo et al., 2022). Within this paradigm, public space performance is conceptualized through interrelated dimensions including accessibility, safety and comfort, inclusivity and equity, perceived publicness, and cultural identity, each of which contributes to shaping user experience and spatial outcomes (Heo et al., 2022). This holistic perspective is further reinforced in emerging studies situated within heritage and tourism contexts, where public spaces are not only functional environments but also carriers of cultural meaning and identity. Recent work on vernacular heritage in Al-Balad illustrates how conservation and tourism agendas interact with the everyday use of such spaces (Al-Gurashi and Shatwan, 2024).

Governance, management, and policy frameworks are as consequential as physical design in shaping how public spaces are used, a point developed in the following section (Nasution and Zahrah, 2019; Binzagr, 2020; Nasreldin and Ibrahim, 2024). Regional studies further demonstrate that public space outcomes are highly contingent upon cultural and institutional contexts. Despite similar design interventions, variations in governance models, social norms, and cultural expectations can produce markedly different user experiences and spatial performances (Shafique and Majid, 2020b). This reinforces the argument that public space evaluation must be context-sensitive, accounting for socio-cultural and political conditions rather than relying solely on universal design principles (Susewind, 2020; Heo et al., 2022).

2.2 Publicness as political and socially constructed

Public space is widely conceptualized as a socio-political arena in which diverse publics negotiate meanings, construct identities, and contest power relations (Barreiro and Gonzalez, 2020). Rather than being neutral or purely physical settings, public spaces are understood as socially constructed environments shaped by competing interests, governance practices, and everyday interactions (Zeng et al., 2025). Within this framework, publicness extends beyond mere physical accessibility to encompass symbolic, social, and political dimensions of inclusion and exclusion. In this study, perceived publicness is defined operationally as a user’s sense of being free to occupy and use a space without needing to consume or justify their presence, of belonging within it, and of being able to express identity there. It is measured through items on the right to be present without buying, freedom to stay, sense of belonging, and comfort in expressing identity (Q26–Q35). The literature emphasizes that publicness operates simultaneously as a process, encompassing production, management, and mediation, and as a product, reflected in spatial form, accessibility, and material conditions (Li et al., 2022; Zeng et al., 2025). This perspective foregrounds the role of governance-mediated access, programmatic interventions, and spatial symbolism in determining who can participate in public life and how belonging is experienced within urban environments (Haider, 2018; Susewind, 2020; Thiollet and Assaf, 2021). Studies situated within Gulf and South Asian contexts reveal inherent tensions between cosmopolitan aspirations and localized socio-cultural practices. Public spaces in these regions often embody a duality wherein globalized urban imaginaries coexist with exclusionary practices shaped by cultural norms, governance structures, and socio-political hierarchies (Orbaşlı, 2017; Thiollet and Assaf, 2021). Such dynamics highlight the need to critically interrogate notions of inclusivity and publicness, particularly in rapidly globalizing urban contexts where access and participation remain unevenly distributed.

2.3 Accessibility, safety, and inclusive design

Accessibility and safety are routinely treated as basic conditions for the use of public space, yet literature is divided on how far physical provision alone shapes experience. One position treats barrier-free design, lighting, and surveillance as direct drivers of use and perceived safety; another holds that their effect depends on social and governance conditions, so that similar features can produce different outcomes across settings. This tension frames the comparison developed here (Shafique, 2024). Empirical studies across diverse contexts demonstrate that physical accessibility, manifested through barrier-free pathways, legible wayfinding systems, and diverse seating arrangements, plays a critical role in enabling equitable use (Li et al., 2022). Simultaneously, safety-related factors such as adequate lighting, clear sightlines, and Crime Prevention Through Environmental Design (CPTED) principles significantly influence both actual and perceived security, thereby shaping patterns of use and user satisfaction (Attia, 2016). Further research underscores the relationship between spatial configuration and safety, illustrating how urban form and environmental design can enhance natural surveillance, reduce crime risks, and improve overall perceptions of security in public housing and mixed-use developments (Attia, 2016). These findings reinforce the importance of integrating spatial analytics, such as space syntax, with environmental design strategies to create safer and more inclusive public environments.

2.4 Cultural identity, heritage, and place-making

Cultural identity and heritage play a fundamental role in shaping how public spaces are perceived, experienced, and appropriated by users (Ginzarly et al., 2024). Rather than functioning as neutral settings, public spaces are embedded within cultural narratives, historical trajectories, and socio-political contexts that influence patterns of use, meanings of belonging, and forms of social interaction (Gehl, 2011; Gehl and Svarre, 2013). In this regard, heritage is not merely a backdrop to public space activity but an active determinant of spatial experience and user engagement. In heritage districts, conservation policy, tourism, and local practice can pull in different directions, so that everyday use negotiates inclusion and exclusion within a regulated setting (Thiollet and Assaf, 2021). This condition reflects broader tensions between global urban imaginaries and locally embedded social norms, particularly in rapidly transforming Gulf cities. Similarly, in Lucknow, Hazratganj’s layered Awadhi, religious, and colonial legacies contribute to place attachment and collective memory that shape belonging independently of physical provision (Haider, 2018; Singh, 2019).

In South Asian contexts, religious processions and festivals structure public space use temporally, while Western European settings such as Nottingham rely on programmed seasonal events and institutional management to achieve similar ends (Heo et al., 2022). Across these diverse contexts, heritage and cultural identity emerge as constitutive dimensions of public space experience rather than passive or symbolic elements (Haider, 2018). They actively mediate how accessibility, inclusivity, and publicness are experienced, thereby influencing the effectiveness of design and governance interventions.

2.5 Theoretical framework

This study treats public-space experience as the joint outcome of five constructs: accessibility, safety and comfort, inclusivity and equity, perceived publicness, and place meaning and cultural identity. The first four are functional and experiential. A space that is easy to reach, feels safe, and is open to a range of users is also more likely to be experienced as genuinely public, so these four are expected to correlate positively with one another. Place meaning and cultural identity is treated as a more symbolic dimension that can persist even where functional quality is low; it is therefore expected to relate positively but more weakly to the other four. The analysis tests these expectations through inter-construct correlation and through comparison across the three sites.

Building on these insights, recent research advocates for composite, user-centered frameworks that integrate accessibility, safety and comfort, inclusivity and equity, perceived publicness, and cultural identity as interconnected dimensions of analysis (Shafique and Majid, 2020a; Heo et al., 2022). Such frameworks provide a basis for comparing public-space performance across contexts and direct attention to how management and programming shape user experience (Heo et al., 2022). Because cultural identity intersects all five dimensions (John et al., 2025), the analysis also examines how heritage and governance regimes mediate the relationship between physical provision and perceived experience across the three sites. They also imply that universal design principles require adaptation to local cultural, religious, and governance conditions if they are to produce equitable outcomes (Nasution and Zahrah, 2019).

3 Methods

This study adopted a cross-cultural, quantitative research design to examine user perceptions of inclusivity, accessibility, safety, publicness, and cultural identity in urban public spaces. A structured questionnaire survey was used to collect comparable data across three culturally distinct urban contexts: Nottingham in the United Kingdom, Jeddah in Saudi Arabia, and Lucknow in India. The methodological approach combined case study comparison, standardized survey administration, and statistical validation techniques to ensure that public space experience could be examined systematically across different socio-cultural and governance settings. The following subsections describe the case study selection, questionnaire design, sampling strategy, data collection procedures, analytical methods, and development of the Cross-Cultural Public Space Inclusivity Index.

3.1 Case studies

The selection of case study sites is central to the objectives of this research, which aims to examine user satisfaction and inclusivity in urban public spaces through a cross-cultural lens. The three selected sites are Old Market Square in Nottingham (United Kingdom), Al Balad in Jeddah (Saudi Arabia), and Hazratganj in Lucknow (India), which represent distinct urban, cultural, and socio-spatial contexts that enable a comprehensive comparative analysis. Old Market Square in Nottingham mentioned in Figure 1 represents a formally planned and institutionally managed public space within a Western European context. As a prominent civic square, it is characterized by structured design, high accessibility, and regular programming of public events. In contrast, Al Balad in Jeddah represents a historic urban district shaped by traditional urban morphology, cultural practices, and climatic conditions (Figure 2). Al Balad is embedded within a socio-cultural framework that strongly influences patterns of use, inclusivity, and publicness.

The site is therefore critical for understanding how cultural identity and heritage contribute to the perception and functionality of inclusive public environments. On the other hand, Hazratganj in Lucknow (India) offers a contrasting example of a highly active, mixed-use commercial heritage street within a South Asian urban context (Figure 3). Unlike the formally designed square in Nottingham or the heritage district of Jeddah, Hazratganj reflects a hybrid condition of formal and informal spatial practices. The selection of these three sites is important for several reasons. First, it captures a broad spectrum of global urban conditions by representing developed, developing, and transitional contexts. This diversity helps identify both universal and context-specific factors influencing public space inclusivity (Figure 4). Second, the study addresses a significant gap in literature, as most existing research on public spaces is limited to single-city or single-region analyses. By employing a consistent methodological framework across three culturally distinct settings, this research enables diverse cultural comparison of user perceptions and experiences. In summary, the selected case studies provide a strategically diverse and analytically robust foundation for examining inclusivity in urban public spaces. The combination of comparative analysis and the proposed evaluative index contribute to current public space research and practice.

The three sites are not typologically equivalent, and comparability is not claimed at the level of built form. The comparison is instead between functionally equivalent entities: in each city, the principal centrally located, publicly accessible, heritage-significant, pedestrian-scaled node in which civic and commercial public life concentrates. Sites were selected against this equivalence criterion rather than against morphological similarity. The study sites were selected using a clear set of common criteria. First, all are centrally located, publicly accessible, multi-functional urban spaces used by residents, visitors, workers, and students, functioning as major social and civic nodes within their cities. This functional equivalence enables meaningful comparison of accessibility, safety, inclusivity, and perceived publicness. Second, all three possess strong historical and cultural significance. Old Market Square represents Nottingham’s historic civic centre, Al Balad reflects Hijazi mercantile heritage as a UNESCO World Heritage Site, and Hazratganj embodies the cultural identity of Awadhi Lucknow. This shared heritage value ensures the relevance of cultural identity across all cases. Third, the selection intentionally spans the Global North (UK), the Gulf (Saudi Arabia), and South Asia (India) to facilitate multi-cultural comparison. The diversity of governance systems, cultural norms, and spatial practices was a deliberate feature of the research design; however, it also limits direct like-for-like comparison and is therefore acknowledged when interpreting the findings.

Fourth, the sites share certain comparative characteristics, including central location, pedestrian use, public accessibility, and their role as important civic and commercial nodes within their respective cities. All are walkable, pedestrian-oriented public spaces occupying approximately 1–3 ha and functioning as focal points within surrounding street networks, ensuring consistent survey conditions across sites. The objective spatial characteristics of each site, including pedestrian infrastructure, lighting, seating, surveillance, and management intensity, are documented comparatively in Table 1. Finally, the study deliberately focuses on second tier rather than capital cities. Nottingham, Jeddah, and Lucknow are major regional cultural and commercial centres shaped more by everyday urban life and local cultural identity than by national image-making. This is methodologically important because public space research disproportionately focuses on capital cities, limiting broader generalisability (Addas et al., 2021; Wang et al., 2025).

Table 1. Comparative overview of objective spatial characteristics for the three study sites.

Spatial conditions were derived from planning documents, academic literature, management reports, and heritage authority sources. Where quantitative data were unavailable, qualitative ratings (High/Moderate/Low) were assigned based on available literature, following established comparative public space evaluation approaches (Heo et al., 2022; Addas, 2022). Key sources: Academy of Urbanism (2019), Al Balad Development Company (2023), Architizer (2015), Gustafson Porter + Bowman (2007), Lucknow Development Authority (2023), UNESCO World Heritage Centre (2023), Verma et al. (2024).

The study adopts a critically informed, pragmatic quantitative approach. Rather than asserting positivist cross-cultural equivalence of perceptions, the CCPSII is framed as a culturally situated, perception-based index. Scores reflect user perceptions within their respective socio-cultural and governance contexts, and not as objective or universally comparable measures of inclusivity. In governance-regulated contexts particularly, perceived scores may be subject to social desirability bias, whereby respondents report satisfaction consistent with prevailing cultural expectations rather than unmediated personal experience, a consideration addressed in detail in Section 5.2. The index is therefore best understood as a comparative analytical tool for identifying patterns across contexts, rather than as an absolute ranking of inclusivity.

3.2 Instrument development

The questionnaire was developed through a structured process. Items were adapted from validated instruments in the public space literature, drawing primarily on Heo et al. (2022) for perceived publicness and urban vitality indicators, Li et al. (2022) for publicness and inclusion items, Addas (2022) for accessibility and management-related items, and Haider (2018) and Susewind (2020) for cultural identity and place-meaning constructs. The initial item pool was reviewed for content validity, cultural appropriateness, and linguistic clarity by the research team, which represented all three study contexts. Co-authors based in or originating from India, the United Kingdom, and Saudi Arabia provided region-specific expertise across the South Asian, UK, and Gulf contexts, respectively. Items identified as culturally ambiguous or repetitive were revised or removed. A pilot study was subsequently conducted with 15 participants at each site (n = 45) to assess item clarity, response time, and cultural sensitivity, leading to minor wording refinements before final administration.

The final questionnaire consisted of 39 Likert-scale items distributed across the five constructs and measured on a five-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). Accessibility assessed physical access, wayfinding, and usability; Safety & Comfort examined perceptions of safety, lighting, and cleanliness; Inclusivity & Equity measured equal participation and access; Perceived Publicness evaluated users' sense of freedom and belonging; and Place Meaning & Cultural Identity explored local heritage and place meaning. The full item set and construct allocation are reported in Supplementary Table S2.

Items were assigned to constructs a priori during instrument development. The final construct allocation reported throughout this paper follows the empirical factor structure subsequently identified in the exploratory analysis (Section 4.2.1), which departed from the a priori assignment in a small number of cases. Two safety-related items perceived safety during the day (Q10) and during the evening or night (Q11) loaded on Accessibility rather than Safety & Comfort, suggesting that perceived security at different times of day is experienced as a condition of access in these settings.

To ensure cross-cultural consistency, the questionnaire was administered in English, Arabic, and Hindi. Translation into Arabic and Hindi was carried out by bilingual members of the research team and reviewed collaboratively for conceptual equivalence, linguistic clarity, and cultural appropriateness, with particular attention to context-sensitive constructs such as perceived publicness and belonging. Formal independent back-translation was not undertaken; this is acknowledged as a limitation, and future cross-cultural applications would benefit from a full translation, back-translation, and reconciliation procedure (Brislin, 1970).

3.3 Sampling, data collection, and survey administration

A quota-based intercept sampling approach was used: quotas set the intended demographic composition at each site, and an interval rule governed who was approached within those quotas. At each site, target quotas were set to ensure approximate gender balance and representation across age groups (18–30, 31–45, 46–60, above 60) and user types (residents, visitors, workers, and students). Respondents were classified according to their relationship with the study site as residents, visitors, workers, or students. Frequency of use was also recorded using four categories: first-time use, 2–3 visits, 10–15 visits, and regular use involving more than 15 visits. These variables were recorded to characterise sample composition and familiarity at each site. Participants approaching or occupying the public space were systematically approached at regular spatial intervals, with every third potential respondent invited to participate to reduce clustering bias. Across the three sites, approximately 384 individuals were invited to take part, of whom around 119 declined or returned incomplete questionnaires, yielding 265 valid responses. The overall response rate was approximately 69%, with 86 valid responses from Nottingham, 88 from Jeddah, and 91 from Lucknow. Data was collected through face-to-face surveys conducted on-site at different times of the day (morning, afternoon, evening, and night) and across weekdays and weekends to capture diverse user groups and activity patterns. Data was collected between June 2025 and January 2026 using standardized procedures to ensure consistency and comparability across sites, with careful attention to language clarity, cultural nuances, and ethical considerations.

The questionnaire was researcher-administered at all three sites. Data collection at each site was conducted by regionally based members of the research team who were fluent in the relevant local language: English in Nottingham, Arabic in Jeddah, and Hindi in Lucknow. All researchers were also able to communicate in English, which served as the common working language of the research team. This composition, which also underpinned the instrument development and translation described in Section 3.2, ensured that items were administered with both linguistic accuracy and familiarity with local conventions of address and interaction. Researchers read each item aloud to the respondent and recorded the response directly, a mode adopted to accommodate variation in literacy across the three settings and to enable participation by respondents approached while standing or walking in the study spaces.

A standardized assistance protocol governed researcher conduct during administration. Researchers were permitted to repeat an item verbatim on request and to clarify unfamiliar terms using a pre-agreed glossary prepared in English, Arabic, and Hindi covering the five constructs and the response scale. Researchers were prohibited from paraphrasing items, supplying illustrative examples, indicating or implying a preferred response, commenting on responses given, or completing items on a respondent's behalf. Where a respondent could not answer an item after one verbatim repetition and a glossary clarification, the item was recorded as missing rather than prompted further.

Because administration was distributed across regionally based researchers rather than conducted by a single team, consistency was maintained through a common written protocol covering study purpose and scope, verbal consent, item administration, the assistance rules above, and the handling of refusals. All researchers involved in data collection were briefed on this protocol prior to fieldwork, and the glossary was translated and reconciled collaboratively by the bilingual members of the research team responsible for the questionnaire translation.

3.4 Ethical considerations

This study was designed as a minimal-risk, non-interventional social-science survey examining adults’ perceptions of publicly accessible urban spaces. The procedures were proportionate to the limited foreseeable risk associated with participation. The study involved no experimental manipulation, clinical procedure, deception, biological sampling, or collection of health information or other sensitive personal data. Participation was voluntary, and verbal informed consent was obtained before questionnaire administration. Potential participants were informed of the study’s purpose, the nature of their participation, the voluntary nature of participation, their right to decline individual questions or discontinue the survey without consequence, and the anonymous treatment of their responses. Only individuals aged 18 years and above were recruited. Participants were approached while independently using publicly accessible urban spaces and were not recruited through employment, educational, healthcare, caregiving, or other dependent relationships. No payment, gift, academic credit, or other inducement was offered.

No names, contact details, photographs, precise residential addresses, health information, or other directly identifiable data were collected. Responses were recorded anonymously, stored in password-protected files accessible only to the research team, analysed in aggregate form, and used solely for academic research purposes. The researcher-administered survey proceeded only after a respondent explicitly agreed to participate and was discontinued if the respondent appeared uncertain or indicated that they did not wish to continue. Formal institutional ethical review was not obtained before data collection. This absence of prospective institutional review is acknowledged transparently as a limitation. Nevertheless, the study incorporated safeguards relating to voluntary participation, informed verbal agreement, adult-only recruitment, data minimisation, anonymity, confidentiality, and participants’ right to decline or discontinue participation.

3.5 Analytical strategy

Data analysis was conducted using IBM SPSS Statistics (Version 24) and AMOS (Version 24). The dataset was first screened for missing values, outliers, and normality. Missing data (<5%) were treated using mean substitution, while outliers were assessed using standardized z-scores (±3threshold). Normality was confirmed through skewness and kurtosis values within ± 2. Descriptive statistics were computed, followed by reliability analysis using Cronbach’s alpha (α > 0.70).

To validate the measurement structure and reduce overfitting, the total sample (N = 265) was randomly split into two subsamples (n = 132 and n = 133). Exploratory Factor Analysis (EFA) was conducted on the first subsample using Principal Axis Factoring with Varimax rotation. Factor retention was based on eigenvalues > 1 and loadings > 0.50, with sampling adequacy confirmed through the Kaiser-Meyer-Olkin (KMO ≥ 0.50) and Bartlett’s Test of Sphericity (p < 0.05). Confirmatory Factor Analysis (CFA) was then performed on the second subsample to validate the factor structure. Model fit was assessed using RMSEA (<0.08), CFI (>0.90), and TLI (>0.90) (Jöreskog and Sörbom, 1986). Convergent validity was evaluated using Composite Reliability (CR > 0.70) and Average Variance Extracted (AVE ≥ 0.50), while discriminant validity was confirmed when the square root of AVE exceeded inter-construct correlations. Further analyses included one-way ANOVA to examine differences across sites, followed by Tukey HSD post-hoc tests. Pearson inter-construct correlation analysis was used to examine the strength and pattern of relationships among the five dimensions. It is important to acknowledge that a total sample of N = 265, split into subsamples of approximately 132/133, is at the lower boundary of what is conventionally recommended for multi-group factor analysis with 39 items. Scholars such as MacCallum et al. (1999) recommend sample sizes of at least 200 per subsample for robust factor recovery. While the adequate model fit indices (KMO = 0.88, CFI = 0.92, RMSEA = 0.060) and high Cronbach’s alpha values (0.82–0.88) provide confidence in the measurement model, future research should replicate these findings with larger, site-balanced samples to further strengthen the generalisability of the framework. To examine whether the five constructs were interpreted comparably across the three cultural groups, multi-group measurement invariance was tested in AMOS. Because the per-site subsamples are modest for multi-group estimation on the full 39-item model, a reduced 15-item five-factor model was used for this analysis, retaining the three strongest and most theoretically representative indicators for each construct, while the full 39-item model was retained as the main measurement validation model. Invariance was assessed sequentially at the configural, metric, and scalar levels following established procedures (Putnick and Bornstein, 2016). Model comparisons relied on changes in approximate fit rather than the chi-square difference test alone, with ΔCFI ≤ 0.010 and ΔRMSEA ≤ 0.015 treated as evidence of invariance (Cheung and Rensvold, 2002). Where full scalar invariance was not supported, a partial scalar model was estimated by releasing non-invariant intercepts identified through modification indices. On this basis, cross-cultural comparisons are interpreted as comparisons of contextually situated perceptions supported by configural and metric equivalence rather than as strictly equivalent latent means.

3.6 Objective spatial indicators

To enable systematic comparison between objective spatial conditions and user perceptions, 13 spatial indicators were identified, each corresponding directly to a validated survey item retained in the final measurement model. These indicators collectively document the observable physical and governance conditions underlying the accessibility, safety, seating, lighting, cleanliness, and wayfinding dimensions examined in this study, and are presented alongside the case study descriptions in Table 1. Their selection is consistent with comparative public space evaluation approaches in the literature (Heo et al., 2022; Addas, 2022). Indicators requiring advance analysis beyond the scope of a literature review and observations, such as formal space syntax analysis or systematic pedestrian counts, are acknowledged as a direction for future research.

3.7 Construction of the CCPSII

Finally, a Cross-Cultural Public Space Inclusivity Index (CCPSII) was developed as a composite measure of inclusivity across the three sites. The index was calculated as the unweighted average of five dimensions, Accessibility, Safety & Comfort, Inclusivity & Equity, Perceived Publicness, and Cultural Identity, enabling direct cross-cultural comparison. An unweight approach was adopted to ensure transparency and comparability across diverse socio-cultural contexts, avoiding sample-specific bias associated with weighted methods. While weighting based on factor loadings or regression coefficients may provide a more optimized representation, such approaches are often context-sensitive and may limit comparability. However, equal weighting assumes all dimensions contribute equally, which may not fully reflect their empirical influence. Therefore, the CCPSII should be interpreted as a comparative empirical tool, with future research exploring weighted approaches using techniques such as Structural Equation Modelling (SEM). The index is therefore better interpreted as a comparative analytical tool for identifying patterns across contexts, rather than as an absolute ranking of inclusivity.

4 Results

4.1 Respondent profile

A total of 265 valid responses were collected across the three study sites, comprising 86 from Nottingham, 88 from Jeddah, and 91 from Lucknow. The full respondent profile is reported in Supplementary Table S1. Gender distribution was close to even overall (50.9% male, 49.1% female) and did not differ significantly across sites, χ²(2) = 1.60, p = 0.45. Age composition was likewise consistent across sites, χ²(6) = 1.91, p = 0.93; the largest group at every site was 18–30 (30.2–33.0%), with the above 60 band somewhat smaller in Jeddah (15.9%) than in Nottingham (23.3%) or Lucknow (23.1%).

User-type composition varied descriptively. Visitors accounted for 31.8% of the Jeddah sample, approximately double the proportion recorded in Nottingham (18.6%) and Lucknow (16.5%), while students were most numerous in Nottingham (32.6%), consistent with the presence of two universities in the city centre, and Lucknow returned the highest resident share (34.1%). Workers were evenly represented across all three sites (19.8–22.0%). The omnibus test was not statistically significant, χ²(6) = 9.29, p = 0.158, Cramér's V = 0.13. This test is distributed across twelve cells, however, and is underpowered to detect a difference concentrated in a single category: the visitor cells alone contribute approximately 59% of the total statistic. The descriptive difference is therefore treated as substantive rather than dismissed on the basis of the omnibus test. Frequency of use differed significantly across sites, χ²(6) = 24.40, p < 0.001, Cramér's V = 0.22. Low-frequency users, defined as those reporting a first visit or two to three prior visits, comprised 51.1% of the Jeddah sample, compared with 26.4% in Lucknow and 19.8% in Nottingham. Notably, first-time users were similarly represented at all three sites (8.1–12.5%); the difference arises principally among respondents reporting two to three prior visits, who constituted 38.6% of the Jeddah sample against 17.6% in Lucknow and 11.6% in Nottingham. Correspondingly, regular users reporting more than fifteen visits comprised 39.5% of the Nottingham sample and 37.4% of the Lucknow sample, but only 20.5% at Jeddah.

This composition reflects the differing user populations of the three spaces rather than a sampling failure. Quotas were set to secure representation across user types at each site rather than to equalize proportions between sites, and Al Balad's status as a UNESCO World Heritage site and a focus of Vision 2030 tourism activation makes a higher share of occasional users the expected condition. The distribution is nevertheless consequential for interpretation, since users with limited accumulated exposure to a space may evaluate accessibility, safety, and inclusivity against different reference points than habitual users. This is examined further in Section 5.1. The sample is therefore broadly balanced on gender and age across sites, while user-type and frequency-of-use composition vary in line with the differing user populations of the three spaces.

4.2 Measurement model validation

To validate the measurement structure while ensuring independent cross-validation, the total sample (N = 265) was randomly split into two approximately equal subsamples (Subsample 1: n = 132 for EFA; Subsample 2: n = 133 for CFA).

4.2.1 Exploratory Factor Analysis (EFA)

Exploratory Factor Analysis (EFA) was conducted on the first subsample (n = 132) using Principal Axis Factoring with Varimax rotation. The Kaiser–Meyer–Olkin (KMO) value was 0.88 and Bartlett’s Test of Sphericity was significant (p < 0.001), confirming excellent sampling adequacy. Five factors corresponding to Accessibility, Safety & Comfort, Inclusivity & Equity, Perceived Publicness, and Cultural Identity were extracted, explaining 62.4% of the total variance in the exploratory subsample. All factor loadings exceeded 0.50, with no significant cross-loadings observed. These results confirm the stability and validity of the five-construct measurement structure (Table 2). Cronbach’s alpha values for the five constructs ranged from 0.82 to 0.88, indicating good internal consistency.

Table 2. Results of Exploratory Factor Analysis (EFA) showing factor loadings, eigenvalues, variance explained, and reliability (Cronbach’s α) for the five constructs.

KMO=0.88, Bartlett’s test: p<0.001.

4.2.2 Confirmatory Factor Analysis (CFA)

Confirmatory Factor Analysis (CFA) was performed on the independent second subsample (n = 133) using AMOS to cross-validate the five-factor measurement model derived from the EFA on the first subsample. The analysis was based on the factor structure identified in the exploratory stage (Figure 5). The results indicated a good model fit, with all key indices meeting or exceeding recommended thresholds: χ²/df = 2.30, RMSEA = 0.060, CFI = 0.92, TLI = 0.91, and GFI = 0.88 (Byrne, 2016) (Table 3). Convergent validity was confirmed, as all constructs demonstrated Composite Reliability (CR) values above 0.82 and Average Variance Extracted (AVE) values above 0.50 (Table 4). Discriminant validity was also established, as the square root of the AVE for each construct exceeded the corresponding inter-construct correlations. Overall, the measurement model demonstrated strong reliability and validity on an independent subsample, supporting its suitability for subsequent statistical analyses including ANOVA, and correlation.

Table 3. Confirmatory Factor Analysis (CFA) model fit indices showing goodness-of-fit statistics for the measurement model.

Table 4. Discriminant validity matrix (square root of AVE) with convergent validity indicators (composite reliability and average variance extracted) for all constructs.

4.3 Measurement invariance across sites

Because the study compares user perceptions across Nottingham, Jeddah, and Lucknow, multi-group measurement invariance was examined to test whether the five constructs were understood comparably across sites. Given the modest site-level sample sizes, a reduced 15-item five-factor model was used as a robustness check, retaining three strong indicators for each construct: Accessibility (Q4, Q5, and Q7), Safety & Comfort (Q13, Q14, Q15), Inclusivity & Equity (Q20, Q21, Q23), Perceived Publicness (Q28, Q29, Q30), and Place Meaning & Cultural Identity (Q37, Q38, Q39) (Figure 6). The full 39-item model was retained as the main measurement validation model.

In the reduced model, all retained indicators loaded meaningfully on their respective constructs, with standardized loadings ranging from 0.65 to 0.82. The standardized correlations among the five latent constructs were positive and moderate to strong, ranging from 0.45 to 0.68, with the strongest association between Accessibility and Safety & Comfort. All inter-construct correlations remained below the threshold at which discriminant validity concerns would normally arise, indicating that the constructs were related but empirically distinguishable.

The configural model, which specified the same five-factor structure across the three groups without equality constraints, demonstrated acceptable fit (χ²/df = 1.80, CFI = 0.90, TLI = 0.914, RMSEA = 0.055, SRMR = 0.052), indicating that the same factor structure was broadly supported across Nottingham, Jeddah, and Lucknow. Constraining the factor loadings to equality in the metric model produced negligible change in fit (ΔCFI = 0.002, ΔRMSEA = 0.001), supporting metric invariance and indicating that respondents across the three sites interpreted the retained indicators in a broadly comparable manner. Full scalar invariance, which additionally constrained item intercepts, was not fully supported, as the change in fit exceeded the recommended threshold (ΔCFI = 0.013). A partial scalar model was therefore estimated by releasing three non-invariant intercepts identified through modification indices (Q13, Q28, and Q37), which produced acceptable fit and an acceptable change in fit relative to the metric model (ΔCFI = 0.007, ΔRMSEA = 0.001). The full set of invariance results is reported in Table 5.

Table 5. Multi-group measurement invariance results for the reduced 15-item five-factor model across Nottingham, Jeddah, and Lucknow.

The partial scalar model was estimated by releasing non-invariant intercepts for Q13, Q28, and Q37, identified through modification indices. ΔCFI and ΔRMSEA are reported relative to the preceding nested model. Invariance was supported where ΔCFI ≤ 0.010 and ΔRMSEA ≤ 0.015 (Cheung and Rensvold, 2002). χ²/df ranged from 1.76 to 1.83 across the four models.

These results provide evidence for configural and metric invariance, with partial support for scalar invariance, across the three cultural contexts. The non-invariant intercepts are substantively interpretable: perceptions of protection from traffic conflict (Q13), sense of belonging (Q28), and learning city culture here (Q37) may carry different baseline meanings across the three urban settings. Accordingly, cross-site comparisons of construct means are interpreted as comparisons of culturally situated perceptions supported by metric equivalence, rather than as strictly equivalent latent mean differences.

4.4 Descriptive statistics and cross-site patterns (RQ1)

To address RQ1, which examines differences in user perceptions across the three study sites, descriptive statistics and one-way ANOVA were conducted. Construct-level mean scores showed distinct cross-cultural patterns in Table 6 and Figure 7. Jeddah recorded the highest scores in Safety & Comfort (M = 4.28), Inclusivity & Equity (M = 4.00), Perceived Publicness (M = 4.00), and Cultural Identity (M = 4.23). Nottingham demonstrated the highest score in Accessibility (M = 3.77), reflecting strong infrastructure and connectivity. In contrast, Lucknow showed comparatively lower scores, particularly in Safety & Comfort (M = 2.63) and Inclusivity & Equity (M = 3.11), although Cultural Identity remained consistently high across all sites (Figure 7).

Table 6. Descriptive statistics (mean and ± standard deviation) of user perceptions across five constructs.

Because social desirability was not directly measured, these scores should be read with caution given the potential influence of cultural response styles and governance contexts. They are interpreted alongside the objective spatial evidence in Table 1, and possible explanations are discussed in Section 5.2.

Table 7 presents item-level mean scores for selected spatial quality indicators across the three sites, providing a more granular examination of the construct-level patterns reported in Table 6. Each indicator corresponds directly to its objective spatial counterpart documented in Table 1, enabling comparison between physical provision and user perception for the same spatial attributes.

Table 7. Item-level mean scores for selected spatial quality indicators across the three study sites.

Indicators 1–11 (Q1–Q11) constitute the Accessibility construct, while indicators 12–13 (Q12 and Q14) are selected Safety and Comfort items included to permit direct comparison with the objective spatial conditions documented in Table 1. Item-level means are presented to one decimal place, whereas the construct-level means and standard deviations in Table 6 were calculated from unrounded respondent-level composite scores. Minor differences may therefore occur when the displayed item means are averaged. All items were measured on a five-point Likert scale (1 = strongly disagree, 5 = strongly agree). Spatial dimension labels reflect the objective physical attribute addressed by each survey item, enabling direct comparison with the documentary spatial evidence presented in Table 1.

A one-way ANOVA was conducted to examine whether differences across sites were statistically significant, as shown in Table 8. The results indicated statistically significant differences across sites for Accessibility (F(2,262) = 6.85, p = 0.001, η² = 0.060), Safety & Comfort (F(2,262) = 8.42, p < 0.001, η² = 0.060), Inclusivity & Equity (F(2,262) = 7.15, p = 0.001, η² = 0.062), and Perceived Publicness (F(2,262) = 5.98, p = 0.003, η² = 0.044). These effect sizes indicate small to moderate differences across sites, with the strongest effects observed for Safety & Comfort. In contrast, differences in Cultural Identity were not statistically significant F(2,262) = 2.45, p = 0.089, η² = 0.018, suggesting that all sites are perceived as culturally meaningful despite other variations. Post-hoc Tukey HSD comparisons further revealed that Nottingham significantly outperformed Lucknow across multiple dimensions, while Jeddah and Nottingham showed no statistically significant difference in Place Meaning & Cultural Identity (p = 0.080), consistent with the non-significant overall ANOVA result for this dimension (Table 9). These results indicate significant variation across the study sites.

Table 8. One-way ANOVA results showing differences in user perceptions across the three study sites, including F-values, degrees of freedom, significance levels (p-values), and effect sizes (η²), indicating the magnitude of variation across constructs.

Table 9. Tukey HSD post-hoc comparisons illustrating pairwise differences between Nottingham, Jeddah, and Lucknow across selected constructs, including mean differences, significance levels, and direction of differences.

Nottingham (N), Jeddah (J), and Lucknow (L).

4.5 Inter-construct correlation analysis (RQ2)

Pearson correlation analysis was conducted to examine the interrelationships among the five public space dimensions across the pooled sample (Table 10). All correlations were positive and statistically significant (p < 0.01), confirming that the five dimensions are meaningfully interrelated and collectively constitute the user experience of public space quality.

Table 10. Pearson correlation matrix showing relationships among key constructs.

**indicates that the correlation is significant at the 0.01 level, two-tailed.

The strongest inter-construct association was observed between Accessibility and Safety & Comfort (r = 0.68), indicating that spaces perceived as physically accessible tend also to be experienced as safer and more comfortable. Perceived Publicness showed moderate-to-strong associations with both Accessibility (r = 0.65) and Safety & Comfort (r = 0.63), suggesting that users' sense of belonging and freedom within a space is closely tied to both its physical openness and its perceived safety. Inclusivity & Equity demonstrated moderate-to-strong correlations with all other dimensions (r = 0.58–0.66), reinforcing its role as a cross-cutting dimension of public space experience. Place Meaning & Cultural Identity, while positively correlated with all dimensions, showed the weakest associations overall (r = 0.52–0.61), suggesting that cultural attachment operates somewhat independently from functional and perceptual dimensions of space quality.

To address RQ2 directly, the mean correlation of each dimension with the other four was computed from Table 10. Safety & Comfort returned the highest mean inter-construct correlation (mean r = 0.63), followed by Accessibility (0.62), Inclusivity & Equity (0.62), Perceived Publicness (0.61), and Place Meaning & Cultural Identity (0.56). No single dimension is therefore uniquely central: the four functional and experiential dimensions form a closely linked cluster, while cultural identity is comparatively independent. These associations are correlational and do not, on their own, indicate causal relationships between dimensions.

4.6 Cross-Cultural Public Space Inclusivity Index (CCPSII) (RQ3)

The Cross-Cultural Public Space Inclusivity Index (CCPSII) reveals variation in public space performance across the three sites (Table 11). Jeddah recorded the highest overall score (3.89), followed by Nottingham (3.55) and Lucknow (3.21). Jeddah’s performance is supported by higher scores in Safety & Comfort, Perceived Publicness, and Inclusivity & Equity, while Nottingham demonstrates strength in Accessibility. In contrast, Lucknow’s lower score is associated with comparatively weaker perceptions of safety and inclusivity despite strong cultural identity. High CCPSII scores in governance-regulated contexts such as Jeddah should be interpreted cautiously, given the potential influence of cultural response styles and governance contexts on how satisfaction is reported. The CCPSII should therefore be understood as a culturally situated, perception-based comparative analytical tool rather than an objective or universally equivalent ranking of inclusivity, with scores reflecting user perceptions within their respective socio-cultural and governance contexts. These considerations are addressed in detail in Sections 5.1 and 5.2, where objective spatial evidence from Table 1 is used to contextualize and interrogate the pattern of perception scores reported here.

Table 11. Cross-Cultural Public Space Inclusivity Index (CCPSII).

5 Discussion

The findings contribute to ongoing debates in public space research by demonstrating that public space performance is not solely determined by physical design but is fundamentally shaped by culturally mediated perceptions of publicness, safety, and inclusion. While traditional approaches emphasize spatial configuration and infrastructure, the findings reveal significant variation in public space experience across Nottingham, Jeddah, and Lucknow. This supports prior research (de León et al., 2020; Addas and Maghrabi, 2021), highlighting that public space experience extends beyond material conditions and is strongly influenced by psychosocial factors such as belonging and social cohesion. The results further reinforce that public space performance is context-dependent rather than universally transferable (Addas, 2022).

It is important to clarify at the outset that the quantitative and critical interpretive dimensions of this study operate at different but complementary analytical levels. The EFA/CFA validation, ANOVA, and CCPSII operate at the level of user perceptions as experienced and expressed within each socio-cultural and governance context. The critical discussion in Sections 5.1 and 5.2 does not invalidate these quantitative patterns but contextualizes them: high perceived inclusivity in Jeddah may simultaneously reflect genuine user satisfaction within culturally mediated norms and fail to capture the experiences of all user groups equally. These perspectives are not contradictory but together constitute a more complete account of public space experience than either approach alone could provide. Accordingly, the objective spatial evidence documented in Table 1 is drawn upon throughout this discussion to distinguish between scores that reflect genuine differences in physical provision and scores that may be mediated by social, cultural, or governance factors.

The comparative analysis reveals distinct patterns across the three contexts. Jeddah records the highest CCPSII composite score, though this finding requires critical interpretation. High scores in safety, inclusivity, and publicness in a governance-regulated heritage context may partly reflect respondents' alignment with prevailing cultural expectations rather than unmediated user experience. As discussed in the sections below, cultural norms, respondent bias, gendered use patterns, and governance conditions may each inflate perceived satisfaction scores in ways that aggregate quantitative methods cannot fully resolve. With this important caveat in mind, Jeddah's strong performance may be attributed to the alignment between cultural expectations, governance structures, and everyday practices, consistent with findings in Saudi contexts (Addas, 2022). However, such alignment may also obscure exclusionary dynamics affecting marginalized groups, highlighting a limitation of purely quantitative assessment. In contrast, Nottingham performs strongly in Accessibility, reflecting effective planning and infrastructure provision, yet does not achieve the highest overall satisfaction. This suggests that accessibility alone does not guarantee positive user experience and supports critiques of infrastructure-led approaches that overlook social and experiential dimensions (Addas and Maghrabi, 2021). Lucknow's lower performance, particularly in Safety & Comfort and Inclusivity, underscores the importance of experiential factors, while its high Cultural Identity scores indicate that symbolic attachment alone is insufficient to ensure positive public space outcomes.

This divergence, in which Nottingham offers stronger objective provision yet Jeddah records higher perceived satisfaction, also invites explanation beyond governance and response style. Expectation-disconfirmation theory holds that satisfaction reflects the gap between what users expect and what they encounter rather than absolute provision, so that settings with locally calibrated expectations can return high satisfaction despite lower objective quality (Oliver, 1980). Adaptation-level theory makes a related point: users judge a space against context-specific reference points, so the same physical feature is evaluated differently across settings (Helson, 1964). Together, these accounts help explain why Nottingham's stronger infrastructure does not translate into the highest perceived quality, and why Jeddah's high scores need not imply that its spaces are objectively superior. Place attachment provides a complementary mechanism: an emotional and identity-based bond with a familiar setting can sustain satisfaction independently of measured physical quality, which is consistent with Lucknow's high cultural-identity scores and the strong attachment to Al Balad as a heritage site (Lewicka, 2011).

The composition of the Jeddah sample provides further purchase on this interpretation. Low-frequency users comprised 51.1% of respondents at Al Balad, against 26.4% in Lucknow and 19.8% in Nottingham (Supplementary Table S1). The difference lies not among first-time visitors, who were similarly represented across all three sites, but among occasional users reporting two to three prior visits. Such users have formed impressions of a space without accumulating the routine exposure through which its everyday frictions become apparent.

This pattern bears directly on the adaptation-level account advanced above, since frequency of use is the mechanism through which such reference points are formed: accumulated exposure is what establishes an adaptation level. The composition difference reported here therefore does not merely qualify the Jeddah result but provides independent support for the interpretive framework applied throughout this discussion, in which perceived quality reflects the relationship between provision and expectation rather than provision alone.

Although frequency-of-use composition differed across the three study sites, an additional one-way ANOVA conducted for the Jeddah sample found no statistically significant difference in Inclusivity and Equity scores among first-time, occasional, repeated, and regular users, F(3, 84) = 0.50, p = 0.683, η² = 0.018. This finding suggests that Jeddah’s comparatively high Inclusivity and Equity score was not primarily driven by respondents’ frequency of use. However, this additional analysis examined only one construct at one study site. Future research should therefore apply stratified or multivariable analyses across all sites and constructs to assess more comprehensively how frequency of use and familiarity influence perceptions of public-space inclusivity.

A key contribution of this study lies in identifying how the five dimensions of public space experience interrelate across cultural contexts. Correlation analysis shows that all dimensions are positively and significantly associated (p < 0.01), confirming that public space experience emerges from a set of mutually reinforcing attributes rather than isolated variables. The strong association between Accessibility and Safety & Comfort (r= 0.68), along with the links between Perceived Publicness and both Accessibility (r = 0.65) and Safety & Comfort (r = 0.63), suggests that users' sense of belonging is closely tied to both physical openness and perceived safety.

Place Meaning & Cultural Identity, while positively associated with all dimensions (r = 0.52–0.61), shows comparatively weaker relationships, suggesting a more independent role shaped by historical and symbolic factors. Notably, despite high cultural identity scores across all sites, Lucknow records lower safety and inclusivity outcomes. This highlights that cultural attachment alone does not compensate for deficiencies in functional and experiential dimensions.

The development of the Cross-Cultural Public Space Inclusivity Index (CCPSII) represents a methodological contribution by providing a structured framework for comparing public space performance across diverse contexts. However, the index simplifies complex socio-spatial dynamics and should be interpreted as an analytical tool rather than a definitive measure.

5.1 Gender, religion, and differential inclusion across the three contexts

A critical dimension that the aggregate quantitative findings of this study cannot fully capture is the differential experience of public space across gender, religious identity, and cultural dress practices. Cross-cultural comparisons of inclusivity remain incomplete unless they consider for whom, and under what social conditions, public spaces are experienced as genuinely inclusive. Although the study employed broadly gender-balanced respondent groups across all sites, the findings were not disaggregated by gender, religious affiliation, dress practice, user type, or frequency of use, potentially masking the uneven experiences of user groups shaped by social norms, religious regulation, and cultural expectations.

In the Jeddah context, women's mobility and visibility in public space have changed considerably following the reforms introduced under Vision 2030. While dress codes and gender segregation policies have formally relaxed, allowing greater public participation (AL-Ban, 2020), informal cultural expectations continue to shape women's spatial experience. Women adopting Western styles of dress or behavior may still encounter heightened social scrutiny compared to women whose appearance aligns more closely with prevailing local norms. Thus, the high levels of perceived inclusivity recorded in Al Balad may reflect the experiences of users whose identities and behaviors conform to dominant social expectations rather than indicating universally unrestricted publicness (Thiollet and Assaf, 2021).

The reverse comparison is equally important. Although public spaces in Nottingham are formally open and protected under equality legislation, visibly Muslim women wearing hijab or niqab may still experience informal exclusion, verbal harassment, or heightened surveillance within secular Western public environments (Siraj, 2011). Nottingham's strong accessibility performance therefore reflects formal inclusion through infrastructure and spatial openness but may not fully capture the lived experience of religiously marked bodies in public space. This reinforces the argument that accessibility alone is a necessary but insufficient condition for genuine inclusivity.

In the Lucknow context, women's experiences of public space are shaped by the intersection of patriarchal social norms, religious identity, and an increasingly polarized political climate. Although Hazratganj demonstrates strong cultural identity and place attachment rooted in Awadhi and Imambara traditions (Haider, 2018; Susewind, 2020), this does not necessarily translate into equal comfort or freedom for all users. Women, particularly those from religious minority communities, may experience public space as simultaneously culturally meaningful and socially restrictive.

Taken together, these findings indicate that publicness and inclusivity are not universal or uniform conditions, but socially mediated experiences shaped by gender, religion, cultural identity, and normative expectations. While the quantitative framework effectively captures aggregate perceptions, it remains limited in revealing the lived experiences of socially marginalized groups. Future research should therefore adopt more explicitly intersectional and qualitative approaches to better understand the differentiated realities of public space inclusion across diverse cultural contexts. These interpretive considerations are further contextualized by the objective spatial evidence presented in Table 1, which documents observed differences in seating provision, shade, surveillance infrastructure, and maintenance management across the three sites. Nottingham recorded the highest or joint-highest perception scores for eight of the eleven Accessibility indicators, including pedestrian accessibility, mobility and disability access, surface safety, seating provision, vegetation and shade, wayfinding, and sensory-impairment accessibility. This alignment suggests that Nottingham’s stronger physical and accessibility provision is generally recognized by users. However, Nottingham did not record the highest scores for basic public facilities or for daytime and night-time safety, indicating that perceptions remain dimension-specific and may also be shaped by management, patterns of use, and socio-cultural context.

A limitation follows from the typological differences between the three sites. Old Market Square is a civic square, Al Balad a heritage souq district, and Hazratganj a mixed-use commercial street, and users approach each with different purposes. This plausibly loads certain constructs unevenly. Perceived Publicness is measured in part through the right to be present without consuming (Q26); in a civic square this is a live question, whereas in a souq and a commercial street commercial exchange is the ordinary reason for presence, so the item may be evaluated against a different baseline. This interpretation is consistent with the invariance results reported in Section 4.3, where the intercept for sense of belonging (Q28) was non-invariant across groups. The functional equivalence of the sites supports comparison of contextually situated perceptions but not of typologically equivalent settings.

5.2 Political governance, social desirability, and the limits of perceived publicness

A critical interpretive issue that requires explicit discussion concerns the extent to which perceived publicness, the central explanatory construct of this study, can be taken at face value across contexts with markedly different governance systems and political freedoms. These governance considerations should be read alongside the objective spatial data presented in Table 1, which reveals that Jeddah's high perceived safety and cleanliness scores are broadly consistent with its objectively documented lighting infrastructure, active security management, and institutional maintenance regime suggesting that at least part of the high scores reflect genuine physical quality rather than solely governance-mediated reporting.

The three study sites represent substantially different political and democratic conditions: the United Kingdom operates as a liberal democracy with constitutionally protected civil liberties and freedom of assembly; India functions as a federal parliamentary democracy with strong constitutional guarantees, though subject to localized governance pressures and socio-political tensions; and Saudi Arabia operates within a governance framework where public behavior and spatial practices are influenced by institutional and cultural regulations. These structural differences have direct implications for how users perceive and report publicness, inclusivity, and freedom within urban spaces.

In governance-regulated contexts, social desirability bias presents a significant methodological concern: respondents may rate publicness, freedom, and inclusivity positively not because they freely experience these qualities, but because prevailing social norms, cultural expectations, or political conditions make the expression of dissatisfaction contextually inappropriate or socially inadvisable (Tourangeau and Yan, 2007). This is particularly salient in the Jeddah context, where strong cultural norms around public conduct, gender behavior, and social conformity may lead respondents to report higher satisfaction with publicness and inclusivity than their lived experience would independently suggest. As Susewind (2020) and Thiollet and Assaf (2021) observe, governance-mediated access to public space shapes not only who physically occupies a space, but also how users articulate their sense of belonging and freedom within it. Perceived publicness in such contexts may therefore reflect culturally bounded satisfaction, a sense of fulfilment within accepted norms, rather than the broader, unconditional publicness implied by liberal democratic frameworks.

This interpretive consideration also applies to Nottingham and Lucknow, rather than to Jeddah alone. Nottingham respondents may be influenced by liberal expectations of openness and by a social desirability pull toward endorsing inclusivity and equality, while Lucknow respondents may judge safety and accessibility against local baselines and report satisfaction calibrated to expectation rather than to absolute provision. Cultural response styles and adaptation effects are therefore present in all three settings, and none should be treated as a neutral benchmark against which the others appear biased (Heine et al., 2002).

Several methodological measures were taken to reduce the risk of socially desirable responding. Surveys were administered anonymously, participation was entirely voluntary, and no personal identifiable information was collected. Instruments were administered in participants’ native languages, Arabic in Jeddah, Hindi in Lucknow, and English in Nottingham, to reduce cognitive burden and improve authentic response. Data collection occurred across multiple times of day and on both weekdays and weekends to capture a range of user groups and activity patterns. Nevertheless, it is important to acknowledge that anonymity alone cannot fully neutralize social desirability effects in contexts where political freedoms are structurally constrained and where certain forms of dissatisfaction may remain socially difficult to express openly. This constitutes an inherent limitation of survey-based research in cross-cultural, politically heterogeneous settings.

These considerations suggest that the findings should be understood within their broader political and cultural contexts. Jeddah’s high scores in perceived publicness and inclusivity should be viewed with appropriate contextual caution. Political structures and governance conditions function as mediating forces that shape not only physical access to public space but also the subjective experience and discursive framing of that access. Future research should therefore integrate political and democratic context variables more explicitly into cross-cultural public space frameworks and consider complementary qualitative methods, such as in-depth interviews or ethnographic observation, to better capture forms of spatial experience and dissatisfaction that survey instruments may not adequately reveal.

6 Conclusion

The research examined user experience and inclusivity in urban public spaces through a cross-cultural comparison of Nottingham, Jeddah, and Lucknow using an integrated, user-centered framework. The findings demonstrate that public space performance is context-dependent and cannot be explained by physical design alone. While accessibility remains essential, it does not independently determine user satisfaction. Instead, the functional and experiential dimensions, accessibility, safety and comfort, perceived publicness, and inclusivity and equity, are closely interconnected (mean inter-construct r ≈ 0.62), while place meaning and cultural identity is comparatively independent (mean r ≈ 0.56). This underscores the importance of experiential and social dimensions alongside physical provision.

Comparative analysis reveals distinct patterns across contexts. Jeddah demonstrates the strongest overall performance, reflecting alignment between cultural context and user experience. Nottingham excels in accessibility but shows comparatively lower experiential quality, while Lucknow, despite strong Place Meaning & Cultural Identity, records lower satisfaction due to limitations in safety and inclusivity. These findings indicate that cultural identity alone cannot compensate for deficiencies in functional and social aspects of public space. The findings show that public space experience emerges through the interaction of physical, social, and perceptual factors. The CCPSII provides a structured comparative analytical framework for evaluating public space performance across diverse contexts; however, as a culturally situated, perception-based index, it should be interpreted alongside objective spatial evidence and critical contextual insights rather than as an objective or universally equivalent measure of inclusivity.

From a practical perspective, the findings highlight the need for context-sensitive planning approaches that prioritize safety, inclusivity, and meaningful social engagement, rather than relying solely on infrastructure-led design. The total sample size of N = 265, split into two subsamples for EFA and CFA, remains at the lower boundary of recommended levels for multi-group factor analysis, with a 39-item instrument (MacCallum et al., 1999). While the model fit indices and reliability coefficients support the adequacy of the measurement model, the findings should be interpreted with appropriate caution and replicated with larger, site-balanced samples in future research. Future research should incorporate qualitative and longitudinal approaches to better capture the complexity of public space experience. Overall, the findings suggest that inclusive and effective public spaces require an integrated approach that combines accessibility with safety, social interaction, and cultural relevance. These insights may help guide the development of more inclusive, resilient, and context-responsive urban public spaces in rapidly urbanizing cities.

Generative AI statement

The author(s) declared that Generative AI was used in the creation of this manuscript. AI was used to assist with language editing like grammar and spelling check. The authors take full responsibility for the originality, accuracy, and integrity of the content, including all analyses and conclusions.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/frsc.2026.1868976/full#supplementary-materialhttps://www.frontiersin.org/articles/10.3389/fpubh.2026.1856870/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Jelena Ristić Trajković, University of Belgrade, Serbia

Reviewed by: Bibhu Kalyan Nayak, Manipal University Jaipur, India

Ola K. Esmail, Ain Shams University, Egypt

Hameda Janahi, Qatar University, Qatar