Next Article in Journal
Assessment and Prediction of Grain Production Considering Climate Change and Air Pollution in China
Next Article in Special Issue
The Perspective Projects Promoting Sustainable Mobility by Active Travel to School on the Example of the Southern Poland Region
Previous Article in Journal
Sustainable Production of Biodiesel Using UV Mutagenesis as a Strategy to Enhance the Lipid Productivity in R. mucilaginosa
Previous Article in Special Issue
Adolescents’ School Travel and Unhealthy Snacking: Associations with School Transport Modes, Neighbourhood Deprivation, and Body Weight
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Perceived Walkability and Respective Urban Determinants: Insights from Bologna and Porto

1
Centre for Territory, Environment and Construction (CTAC), University of Minho, 4800-058 Guimarães, Portugal
2
SYSTEMA Research Centre, European University Cyprus, Engomi, Nicosia 2404, Cyprus
3
Alma Mater Studiorum, University of Bologna, 40136 Bologna, Italy
4
CitUpia AB, 104 30 Stockholm, Sweden
*
Author to whom correspondence should be addressed.
Sustainability 2022, 14(15), 9089; https://doi.org/10.3390/su14159089
Submission received: 21 May 2022 / Revised: 9 July 2022 / Accepted: 20 July 2022 / Published: 25 July 2022

Abstract

:
Walking is undoubtedly a sustainable and healthy mode of transport. However, the decision to walk is influenced by many built environment and streetscape attributes. Specifically, the term walkability is used to describe the extent to which the urban environment is pedestrian-friendly, usually by quantifying multiple built environment attributes at the neighbourhood scale. The present study adopts a qualitative approach to evaluate perceived walkability. Based on a questionnaire (n = 1438) administered in the cities of Bologna and Porto, this paper analyses how respondents perceived and evaluated 19 built environment and streetscape attributes. An Exploratory Factor Analysis was carried out to examine the correlations between the various attributes and to identify the underlying walkability determinants. The analysis indicated that 13 attributes were highly correlated, resulting in four determinants: (i) urban ambiance, which includes land use and street design attributes, such as land use mix, enclosure, transparency, and architectural and landscape diversity; (ii) pedestrian infrastructure, which is related to sidewalk conditions; (iii) street connectivity and proximity to community facilities; and iv) access to other modes of transport. In turn, traffic safety and security were not correlated with perceived walkability in both cities. These findings suggest that specific urban design and pedestrian infrastructure attributes should be highly considered when formulating policies aiming to create more pedestrian-friendly cities, as well as in walkability studies and when developing walkability scores and indexes.

1. Introduction

Walking is the most common and natural form of moving and an enjoyable, sustainable, and healthy mode of transport, namely for short urban trips. The overall environmental and health benefits of walking have been extensively analysed and discussed in the literature [1,2,3]. Due to these benefits, many studies have been conducted to analyse how the built environment affects travel behaviour and the overall walking experience [4,5]. The extent to which the built environment is pedestrian-friendly and enables walking is broadly defined as walkability [6]. The concept of walkability only started to be used in transport and planning studies in the early 2000s, but rapidly became a widely used concept in sustainable mobility and other disciplines. Nonetheless, the term is rarely defined in English dictionaries [7], and the concept associated with walkability remains vague and unclear [8]. In part, this is related to the different subject areas that included walking and walkability in their research field, such as transport, urban planning, health, and environmental sciences. For these reasons, the ways to assess walkability are strongly changeable regarding the attributes, the methods, and the scale of analysis used.
Walkability assessments have been mostly based on quantitative evaluations in the form of indexes or metrics that measure a changeable number of built environment attributes [9]. One of these first referential methods was the so-called “3D layout” proposed by Cervero and Kockelman [10]. They found that density, land-use diversity, and pedestrian-oriented designs reduce automobile trips and encourage active travel. Later, Ewing and Cervero [11] added destination accessibility and distance to transit (“5D layout”) to have a greater overview of the influence of built environment features on walking. Other referential methods were the GIS-based walkability indexes of Frank et al. [12,13], which included land use mix, street connectivity, residential density, and retail floor area. Acting as a predictor of walking and physical activity, these composite indexes have been extensively replicated and have inspired many other indexes [6,14]. The Walk Score is another example of a quantitative index used for assessing walkability [15]. The Walk Score uses distance to amenities, block length, and intersection density to estimate walkability.
Walkability research has resulted in increased precision in measuring-built environment influences of walking [16], namely due to the growing availability of spatial data along with the use of GIS tools [17]. For Andrews et al. [18], the widespread use of quantitative approaches is a “neo-environmental determinism”, which assumes a deterministic relationship between the environment measured by specific attributes and walking. However, these objective measures often do not include important qualitative features such as the aesthetics of the landscape, the sense of safety, and the pleasure of walking. In this context, Jensen et al. [19] also emphasised that walkability indexes usually overlook streetscape features, such as the characteristics of sidewalks, which are associated with perceived walkability [20]. Selecting an appropriate scale of analysis to accurately represent daily activities, such as walking trips, also remains a topic of discussion [21]. Scales of analysis vary greatly from country to country depending on the characteristics of cities, while buffers often used in walkability studies based on Euclidian distances from specific points may be poorly suited to describing walking. This leads to another problem which consists of replicating indexes developed for specific areas. As shown by Shashank and Schuurman [21], the application of specific methods may result in different walkability scores for the same city. Various urban morphologies, data in different formats and from different spatial units could make replications less effective in describing walkability.
Finally, the role that social and individual characteristics play in the different perceptions of spaces to walk, such as the socioeconomic status, social cohesion, community identity, race, age, gender, among others are often not taken into account [16,22,23]. All these variables could significantly affect people’s willingness to walk and, therefore, understanding the mechanisms by which the social environment influences walking remains limited [16]. This also explains that objective and perceived walkability do not always correlate in the same direction, and a neighbourhood with high objective walkability does not always correspond to a neighbourhood with high perceived walkability [17]. Moreover, as shown in some studies, people may not walk and socially interact more simply because they live in a walkable area, due to the complex nature of the relationships between walkability and the social environment [24,25].
Thus, research on walkability lacks qualitative studies, which according to Dörrzapf et al. [26] should be part of a new walkability concept and understanding. More specifically, there is a lack of studies on how citizens’ views favour or hinder the propensity to walk [27] and on how pedestrians perceive walkability [28]. In addition, as research has been mostly conducted at the neighbourhood level, there is a lack of studies evaluating both neighbourhood and micro-scale environmental factors when some forms of utilitarian walking are analysed [29].
Understanding how urban space influences citizens’ perceptions of walkability is critical to designing efficient pedestrian planning policies and guiding future research in this field. In this paper, we explore this topic by evaluating the influence of 19 built environment and streetscape attributes on perceived walkability. The evaluation was carried out by performing a questionnaire (n = 1438) in the cities of Bologna (Italy) and Porto (Portugal) in the context of the research project called Smart Pedestrian Net. Respondents were asked to rate the importance of these attributes in walking using a five-point Likert scale. An Exploratory Factor Analysis (EFA) was the method selected to identify correlations between the attributes and to highlight the main determinants associated to walkability. To the best of our knowledge, this is the first study using a qualitative approach to analyse the way people perceive walkability in both cities.

2. Literature Review

Empirical research has shown that walking behaviour is affected by the way people perceive the walking environment [22]. The decision and satisfaction of walking are influenced by multiple built environment and streetscape attributes. One of the most widely used tools to assess individuals’ perception of walkability is the Neighbourhood Environment Walkability Survey (NEWS) developed by Saelens et al. [30]. NEWS covers eight main walkability dimensions: residential density, land use mix, land use mix access, street connectivity, infrastructure for walking, traffic safety, security from crime, and aesthetics. In a recent review on the influence of built environment and streetscape attributes on walkability, Fonseca et al. [5] identified seven dimensions: land-use density, land-use diversity, accessibility, street network connectivity, pedestrian facilities and comfort, safety/security, and streetscape design. Accordingly, the influence of these built environment and streetscape dimensions on walking is briefly analysed below.
Land-use density and particularly residential density are crucial dimensions of the city and its built environment. High residential density means a larger number of people living in a given area, which potentially leads to an increase in walking activities. More people walking will encourage others to walk, increasing the interactions that take place between people [31]. In turn, traffic congestion tends to increase with population density, which could make walking more convenient rather than driving [32]. Areas with high residential density also tend to attract various amenities, which reduces travel distance and time, making walking trips more convenient. For these reasons, previous studies have shown that residential density is associated with more walking time and more active people [12,33,34].
Land-use diversity shows the extent to which there is a mix of land uses within an area. An area with a mix of land uses typically offers more non-residential destinations (services, shops, restaurants, etc.), which generally reduces the distances to travel and makes walking more convenient. Land-use diversity has often been evaluated by entropy indexes showing the degree to which an area has uniformly diverse land use through normalised scores, where 0 means a single-use and 1 an even distribution of a specific number of land uses [12,35]. Alternative ways of measuring diversity have included: (i) the number of local destinations in an area [36]; (ii) the percentage of specific land use in an area [14]; and (iii) the retail floor area, which indicates the amount of available space for parking reflecting a higher propensity to car usage [13]. The literature shows that mixed land uses are more conducive to walking and are associated with more active people [12,35].
Accessibility is a complex and context-dependent concept that has been used to describe ease of access in terms of distance to attractions, but also as the expected utility derived from the number of accessible amenities (services, urban functions) from a given area [11,37]. In this study, accessibility reflects the walking distance to key destinations and public transport. Travel distance is critical for pedestrians because walking is a low-speed mode and requires more time and physical effort to reach distant destinations. For that reason, as the travelling distance increases, the probability of using motorised transport also increases [38]. Some authors argued that neighbourhood amenities, such as retail, stores, and services, should be provided within around 15-min walking or cycling distances [39]. Compact urban areas with mixed land uses are known for reducing travel distances, which increases the propensity to walk. Distance to public transport is also a critical attribute. There is evidence that the shorter the distances to stops, the higher the odds of walking to catch a bus or a train [40]. The literature shows that people are usually willing to walk around 400 m to arrive at bus stops and 800 m to arrive at rail stations [40,41].
Street connectivity can be understood as the directness and availability of alternative routes between destinations. More interconnected streets provide more potential routes for walking, which reduces distances between origins and destinations and makes walking more convenient [40]. Street connectivity has been described by a considerable number of different attributes such as intersection density, street density, cul-de-sac density, average block length, median block length, connected node ratio, link node ratio, among others. From these, intersection density, usually calculated as the number of intersections of three or more links in an area, has been the most widely used attribute to describe street connectivity [5]. Areas providing high intersection density have been correlated with more walking activity and physical activity [12,42].
Pedestrian facilities are infrastructure provided to enhance the comfort and safety of pedestrians. They include infrastructure such as sidewalks, crosswalks, underpasses and overpasses, street furniture, and traffic calming devices, among others. In general, research has shown that well-designed and maintained pedestrian facilities encourage walking and promote higher levels of pedestrian travel. Sidewalks are the basic pedestrian facility, but attractive sidewalks depend on various features. Previous research has shown that the presence of wide, well-maintained, and clean sidewalks encourages pedestrian activity [43,44]. The presence of street furniture helps to create more pleasant sidewalks, providing for example settings for resting, but the presence of obstacles, such as parked cars on sidewalks, reduces the area available for walking and discourages people from walking [4,43]. Street trees provide various benefits at the sidewalk level (shading, cooling, UV protection, traffic safety, and enclosure, among others). For these reasons, street trees have been positively associated with healthy pedestrian routes [45]. Slopes have a significant influence on walking since they affect the travel speed and the effort required for walking [45]. Sloped sidewalks (>5%) are generally considered unattractive for walking [46].
Traffic safety means that pedestrians should be protected from motorised traffic [15], while security means that pedestrians should be protected from crime and incivilities [47]. As pedestrians are vulnerable road users, traffic safety is a critical issue. Some of the most commonly used attributes to describe traffic safety include traffic speed, traffic volume, the number of traffic lanes, among others. It has been shown that high traffic volume is a barrier to walking [14,48], the risk of accidents was associated with less physical activity [49], and pedestrians prefer quiet streets with low traffic speeds and volumes [43] and streets with few lanes to cross [50]. In turn, public security has been reported as a strong deterrent to walking [47,51], particularly in regions characterised by urban violence [4,47]. Security has been measured by considering various attributes, such as street lighting, buildings with broken windows and graffiti, homicide rates, police officers/stations, volume of pedestrians, among others [5]. It has been shown that graffiti, few shops, and vacant buildings increase the perception of insecurity [51], while the presence of street lighting, surveillance systems and police officers enhance the perception of security [4].
Streetscape design reflects the extent to which pedestrian spaces are attractive and functional for walking. Streetscape design is related to street-level features, such as the presence of attractive pedestrian facilities and furniture [5]. The literature shows that street design features have an overall positive impact on walking [52,53] and that they have been positively associated with walking and physical activity [54]. The attributes most commonly used to assess streetscape design include: (i) human scale, reflecting how buildings and spaces are scaled to human size and needs [53]; (ii) enclosure, showing how enclosed by vertical elements the spaces are, creating a room-like effect [52]; (iii) complexity, indicating the visual richness of a place, in terms of buildings shapes, styles, colours, street furniture, among others [52]; and (iv) transparency, showing the proportion of windows and doors at the street level having active uses [52].
To sum up, this review shows that the concept of walkability and the way people perceive walkability are influenced by a significant number of attributes. In this study, we intended to understand which urban determinants may explain walkability according to the perceptions of a group of individuals from Bologna and Porto.

3. Method and Data

This section describes the methodology used to collect and analyse the built environment and streetscape data in Bologna and Porto.

3.1. Study Areas

This study was conducted within the research project called Smart Pedestrian Net (SPN). The main goal of SPN was to increase walkability and to encourage people to walk on a daily basis in Bologna and Porto. Bologna and Porto are two medium-sized southern European cities (Figure 1).
Founded in the Middle Ages, they are recognised for their rich history and heritage. Bologna is particularly famous for its monuments and extensive porticoes that cover most of the city centre, while the historical centre of Porto has been classified as a World Heritage Site since 1996. Both cities are engaged in promoting sustainable mobility policies to limit the negative effects of motorised traffic, especially in the city centres. This includes policies that have led to an increase in electric mobility, public transport, and active modes of transport. Both cities are engaged in improving walkability through various measures such as the provision of pedestrian-only streets, the restriction of motorised traffic in some streets, and the adoption of measures to improve pedestrian facilities and traffic safety, among others.

3.2. Data Collection

Based on an extensive literature review on the influence of built environment and streetscape attributes on walkability [5], a total of 19 attributes were selected to evaluate their influence on individuals’ perception of walkability (Table 1).
The evaluation of the 19 attributes was carried out by performing a questionnaire (n = 1438) in the cities of Bologna and Porto. The questionnaire was structured on Google Forms and consisted of a mix of single choice, multiple-choice, ranking, and open-ended questions divided into four main parts. The first aimed at collecting personal information, such as gender, age, education level, and type of activity. The second and the fourth parts were to collect data about the frequency and purpose of walking [55] and the use of pedestrian navigation apps [56], respectively. The third part was to evaluate the 19 attributes presented in Table 1. Participants were asked to rank the importance of each attribute when walking by using a five-point Likert scale, ranging from 1 (“not important”) to 5 (“very important”).
A description of the meaning of each attribute was included in the questionnaire as follows: “close proximity” was defined as bus stops and train stations at a walking distance up to 400 and 800 m, respectively, car parking at 500 m and community facilities at 1000 m; “street connectivity” was explained as the availability of alternative routes (street intersections) between destinations; “low traffic speed” as the streets where maximum limit is 30 km/h; streets with few lanes as those having ≤2 traffic lanes; “wide sidewalks” as those providing ≥1.50 m free of obstacles; “sidewalks in good condition” were those without major deformities (cracks, holes, raised pavements) that may cause trip hazards; “unobstructed sidewalks” as those without physical obstacles, such as parked cars; “street furniture” as objects and equipment installed along sidewalks, such as benches and litterbins; “low slopes” as sidewalks having a longitudinal gradient ≤2%; “sidewalks with trees/greenery” as those containing trees in segments of at least 10 m; “streets with many pedestrians” as those having at least some pedestrian movement/visible pedestrians walking; “shopping streets/areas” as commercial streets having different stores; “high residential density” as areas with residential buildings and more people living there; “mixed land uses” as areas containing diverse uses and functions, such as residential, retail, and recreational; “enclosure” as the extent to which streets are visually defined by buildings, walls, trees, and other vertical elements; “architectural and landscape diversity” as the visual richness of a street, namely in terms of building colours, architectural styles, and outdoor dining; and “transparency” as the proportion of windows and doors at the street level having active uses.
After consolidating the structure of the questionnaire, a pilot test was performed to gauge the various questions and their overall organisation. As a result, some changes were carried out to improve the reliability and sensitivity of the questionnaire.
The second step of the work consisted of obtaining a statistically significant sample of the population living in both cities. The widely used Cochran’s formula [57] was adopted to calculate the sample size in each city. The formula is shown in Equation (1).
n = z 2 p q e 2 1 + 1 N   ( z 2 p q e 2 1 )
where: n is the sample size, N is the population size, z is the critical value (1.96) for the 95% confidence level, p is the sample proportion (0.5), q is equal to 0.5 (q = 1 − p), and e is the margin of error (0.05). Considering that the population living in Bologna and in Porto in 2019 was 301,984 inhabitants [58] and 216,606 inhabitants [59] respectively, a sample of 384 individuals in each city was required.
The third step of the work was the administration of the questionnaire. It was decided to administer the questionnaire electronically, because online surveys open up the survey to more people, they allow the respondent to complete the survey at a convenient time and at a pace that is comfortable for the respondent, and because they encourage the participation of people who have more intense attitudes regarding the asked topic [60]. Respondents were approached through the SPN website, databases from the municipalities and the universities of Bologna and Porto and through social media. In Bologna, the questionnaire was distributed in Italian from May to July 2019; in Porto, the questionnaire was distributed in Portuguese from September to November 2019. Thus, both questionnaires were administered a few months before the COVID-19 pandemic, which severely impacted urban mobility.

3.3. Data Analysis

A descriptive analysis was performed to characterise the sample and to analyse how the attributes were evaluated in terms of mean, median, and standard deviation. Statistically, the study was based on an EFA, which refers to a set of statistical techniques often used to identify the underlying factors measured by a group of observed and interrelated variables. Factor analyses of Likert questions with five or more scale points were adopted in previous transport studies [61,62]. The principal component analysis, which measures the variance in the set of variables, was the EFA method adopted. Three main steps were developed to perform the factor analysis: (i) check the suitability of data; (ii) factor extraction; and (iii) factor rotation and interpretation.
Regarding the first step, the statistical measure used to check the suitability of data was the Kaiser–Meyer–Olkin (KMO) method. The Kaiser measure is a ratio that represents the amount of the total variance explained by each factor. KMO values between 0.8 to 1.0 indicate that the sampling is adequate, while values between 0.7 to 0.8 are still acceptable [62]. In addition, the Bartlett’s Test of Sphericity evaluates the overall significance of the correlation matrix, by testing the null hypothesis, e.g., if there is or if there is no correlation between the variables. If there is no correlation, the variables are unsuitable for structure detection. The significant value p < 0.05 shows that factor analysis may be worthwhile for the data set.
Regarding the second step, the factor extraction consists of determining the number of factors that can be used to represent the correlations among the variables. The method used was the Principal Component Analysis through the Kaiser’s Criterion and the Scree Test. The Kaiser’s criterion corresponds to a score called eigenvalue that represents the amount of the total variance explained by each factor (or component). Only components with high eigenvalues are likely to represent real underlying factors. As adopted in previous studies [62,63], we only selected components having eigenvalues above 1.0, because they are considered to be significant. These components will be used to identify the main determinants associated with walkability. The Scree Test was also used to help to extract the underlying factors. The Scree plot displays the components’ eigenvalues versus the total number of components. Eigenvalues are plotted as dots within the graph, and a line connects successive values. By stopping at the point where this is a levelling at the plot area, this test can be used to identify the optimum number of components that can be extracted [63], The so-called communalities (r-square values) show the extent to which the underlying factors account for the variance of all the variables.
The factors obtained in the initial extraction could be difficult to interpret due to cross-loadings as some variables could have more than one-factor loading. The factor rotation and interpretation, performed during the third step, solve this problem by redistributing the factor loadings over the factors according to some mathematical rules. In this study, a varimax rotation, which is among the most recommended and widely used methods [64], was performed to achieve a meaningful factor matrix and to understand how the factors correlated with each component. In a varimax rotation, the correlation of each of the variables is the closest to 1.0 with only one of the factors obtained, and close to zero with all other factors identified. The Statistical Package for the Social Sciences (SPSS) was the software used to perform the described factor analysis.
Finally, the reliability coefficient Cronbach’s alpha was utilised to assess the internal consistency of the obtained factors. This coefficient has been used to measure the reliability of the questionnaire with multiple Likert scale questions [65]. Cronbach’s alpha values of 0.70 or higher indicate good reliability of data [63,65,66].

4. Results

4.1. Sample Description

A total of 1438 individuals (60% from Bologna and 40% from Porto) evaluated the 19 attributes. As shown in Table 2, in both cities respondents included slightly more females, aged between 45–65 and 25–45 years old, who had an undergraduate degree, were living in the respective cities, and were employed full-time.
The sample’s sociodemographic characteristics do not differ much from the population of each city in some variables (gender and people living in each city), but there are deviations in some variables. For example, elderly people (≥65 years old) are underrepresented, while adults, the employed, and graduates are overrepresented. These deviations are mostly explained by the difficulty in targeting specific groups with online questionnaires.

4.2. Sample Description

Table 3 shows some descriptive results from the evaluation made by the participants. In general, the attributes related to pedestrian facilities were those that were highly scored (mean > 4.0) and those with the lowest standard deviation values (<1.0). In turn, the less scored attributes (mean < 3.0) include some attributes such as: “walk in areas closer to car parking”, “walk in streets with ≤2 traffic lanes”, “walk in areas with high residential density”, and “walk in streets providing enclosure”.

4.3. Factor Analysis

To ensure that the data collected was suitable and adequate for EFA, the KMO measure of sampling adequacy and Bartlett’s Test of Sphericity were firstly checked (Table 4). The KMO measures the factorability of the variables individually and as a group, which is a basic assumption for factor analysis. The obtained measure of sampling adequacy of 0.902 falls in the range between 0.8 to 1.0, which is considered adequate. In turn, the Bartlett’s Test of Sphericity was significant (p < 0.001), indicating that the correlations between the variables when taken collectively are significant at one per cent level, e.g., there was a substantial correlation in the data. Thus, the KMO and the Bartlett’s test indicated that the outcomes of the factor analysis were meaningful.
The extraction scores of the 19 attributes are shown in Table 5. The communalities, which reflect the common variance in the database, show that after the initial extraction, most communality loadings were higher than the minimum cut-off of 0.500. From the 19 attributes, only 6 had a communality lower than 0.500: “walk in streets with ≤2 traffic lanes”; “walk on sidewalks with trees/greenery”; “walk on sidewalks with low slopes”; “walk on sidewalks with street furniture”; “walk in streets with low traffic speed”; and “walk in streets with many pedestrians”.
Then, a Principal Component Analysis was adopted to determine the number of factors that represent the correlations among the attributes. Table 6 shows the eigenvalues and total variance explained resulting from the application of the Kaiser’s criterion. After the initial extraction, it can be observed that most eigenvalues in column 2 were less than the minimum threshold of 1.0 and only four components had an eigenvalue greater than 1.0. The dominant eigenvalues in decreasing order for Components 1–4 are 6.29, 2.34, 1.46, and 1.06, respectively. In terms of variance, Components 1–4 explain around 33%, 12%, 8%, and 6%, respectively of the total variance. Thus, 59% of the total variance is explained by these four components.
The Scree Test (Figure 2) also confirms that there is a remarkable variation in slope between the 19 components. The first four components explain most of the variability, while the remaining 15 components explain a much lesser proportion of the variability and are likely to be less important. Thus, the Scree Test confirmed the Kaiser’s criterion, indicating that four factors could be a realistic number to represent the determinants that influence the perceptions of walkability in both cities.
However, the components resulting from the initial extraction do not show which specific attributes are contributing more to the total variance. Thus, a varimax rotation was performed to check how these variables are correlated and how they can be reduced to a smaller number of main determinants. These results are shown in Table 7.
After rotating the components, the cumulative proportion of the described variance did not change as it was kept at 59%, but the variance credited to the components did (Table 6). After rotation, the variance of Component 1 decreased to 21.9%, while the variance of Components 2–4 rose to 16.8%, 10.4% and 9.5% respectively.
Based on the distribution of the attributes by the components, it can be concluded that, for the participants in this study, four main determinants influence their perceptions of walkability. We labelled these determinants as: (i) urban ambiance; (ii) pedestrian infrastructure; (iii) connectivity and community facilities; and (iv) access to other modes of transport. Component 1, urban ambiance, has six attributes related to land use and urban design. Component 2, pedestrian infrastructure, is defined by three streetscape attributes related to pedestrian facilities. Component 3, connectivity and community facilities, consists of two attributes associated with the structure of the street network and to the proximity to public facilities. Finally, Component 4 includes two attributes related to access to other modes of transport.
As mentioned in the Methodology, Cronbach’s alpha was utilised to assess the internal consistency of the results. The Cronbach’s alpha showed acceptable values for an exploratory study and a good reliability level (0.83). As shown in Table 7, the components of urban ambiance, pedestrian infrastructure and connectivity, and community facilities have Cronbach’s alpha values of 0.85, 0.79, and 0.70, respectively. The component access to other modes of transport has a slightly lower value, which is still close to the threshold of 0.70. The Cronbach’s alpha values confirmed the reliability of the questionnaire and attested to the confidence of the results.

5. Discussion

According to the results, four main determinants were found to be associated with perceived walkability: urban ambiance, pedestrian infrastructure, connectivity, and community facilities, and access to other modes of transport (Figure 3). These four determinants account for 59% of the total variance, fulfilling the minimum cumulative proportion of variance threshold of at least 50% [67].
Urban ambiance is a concept that goes beyond the spatial attributes of spaces as it also involves a socio-aesthetic approach that enables people to grasp everyday urban atmospheres [68]. Determinant 1 was denominated urban ambiance precisely for mixing tangible and intangible attributes and bringing material and affective dimensions to the built environment. “Walk in areas with high residential density” was the attribute with the highest correlation (0.823) with urban ambiance. This finding shows that areas with high residential densities are more conducive for walking than areas with low densities, confirming previous studies on the association of residential density with higher pedestrian movement, walking time, and people being physically active [12,33,34]. “Walk in shopping streets/areas” obtained a correlation of 0.743 with urban ambiance. This is not surprising since Bologna and Porto, mainly their central areas, contain many commercial streets. Confirming previous research [69], this study shows that shopping is a main daily trip purpose and that commercial streets are attractive for pedestrians since they fulfil various daily needs. “Walk in areas with mixed land uses” obtained a similar loading value (0.729). Providing diverse uses (residential, commercial, services, recreational) ensures that residents can benefit from essentials within their residential areas, reducing travel distances and the need of using motorised transport. Therefore, this finding also confirms previous research showing that areas with mixed uses are more pedestrian-friendly [6,12].
The analysis also indicated that some street design features are strongly correlated with urban ambiance. They are known for influencing the overall experience and satisfaction of walking, but this influence remains relatively unclear due to the difficulty in assessing these attributes, such as enclosure [70]. In our study, “walk in streets providing enclosure” has a correlation of 0.794 with urban ambiance. This attribute reflects the extent to which pedestrians are encapsulated by the built environment. Enclosure influences the perception of space confinement, creating a sense of intimacy, security, and livability [70]. The high correlation of this attribute with urban ambiance could be explained by the network of 38 km of porticoes that cover the city centre of Bologna, which provide a room-like effect and, therefore, a suitable perception of enclosure. “Walk in streets providing transparency” was also correlated with urban ambiance (0.676). This finding indicates that the degree to which people can see or perceive human activity beyond the edge of a street makes a pedestrian environment more attractive, as facade transparency, understood as the proportion of doors and windows with transparent glass, is a proxy for shopping and security (more eyes on the street). “Walk in streets with architectural and landscape diversity” was also correlated with urban ambiance (0.557). This finding suggests that the variety of buildings shapes, sizes, materials, and colours, the diversity of architectural styles, historical buildings, and ornamentation influences the visual interest of pedestrians and make the urban environments more attractive. This conclusion can also be found in some studies on the impact of urban design features on walkability [71]. The correlation of this attribute with urban ambiance could be explained by the remarkable built heritage found in both cities. Evidence from previous studies demonstrated that people do value the architectural heritage and tend to perceive these characteristics as pleasant [72].
In our study, determinant 2 is related to pedestrian infrastructure and three attributes were correlated with this determinant: (i) “walk on sidewalks in good condition” (0.795); (ii) “walk on unobstructed sidewalks” (0.770); and (iii) “walk in streets with wide sidewalks” (0.730). In general, there is evidence that a suitable pedestrian infrastructure improves the comfort and safety of walking and encourages people to walk [73]. A sidewalk in good condition requires the use of smooth, stable, and slip-resistant paving materials and suitable maintenance over time. The second correlation shows that sidewalks free from obstructions are valued by pedestrians. Obstacles on sidewalks are undesirable as they reduce the space available for walking and force pedestrians to make detours [4]. The third correlation confirms that pedestrians prefer to walk on wide sidewalks. Wide sidewalks provide enough space to walk side-by-side and are less sensitive to pedestrian congestion. This correlation could also be explained by the compact urban structure found in the centres of Bologna and Porto, where many sidewalks are not wide enough for walking comfortably namely at peak hours.
Determinant 3 is related to connectivity and community facilities. These facilities include a wide range of services (educational, health, cultural, recreational, etc.) that meet people’s daily needs. The correlation of the attribute “walk in areas closer to community facilities” (0.717) confirms that distance/proximity to community facilities is critical for pedestrians as found in previous research [14,15]. From a planning perspective, this finding suggests that most urban services and amenities should be within appropriate walkable distances. In turn, “walk in areas with high street connectivity” had a correlation of 0.705 with this factor. For the participants in this study, street connectivity, understood as the availability of alternative routes between destinations, also influences their perceptions of walkability. As found in previous studies, this finding also confirms that well-connected street networks have a significant influence on active travel [42] and on utilitarian trips [48,54].
Finally, determinant 4 comprises two attributes related to the access to other modes of transport, namely car parking (0.806) and public transport stops (0.758). This factor is associated with long multi-stage trips that cannot be done exclusively on foot. As shown in previous studies, an appropriate coverage and density of public transport stops are vital to encourage active travel to public transport [11]. The literature shows that, on average, bus stops should be within a distance of 400 m, while subways and rail stations should be within a walkable distance of 800 m [40]. The attribute “walk in areas closer to car parking” could be explained by the significant share of car trips that are still found in both cities. According to the Census of 2011, more than 50% of the commuting trips to Bologna and Porto were based on individual motorised modes [74,75]. The work of Fonseca et al. [55] on the commuting trips to the central areas of these cities also showed that 47% were multi-stage trips, while 32% were private motorised trips. For these reasons, the proximity to car parking is still seen as an important attribute with an impact on urban mobility.
From the 19 attributes evaluated, six obtained low loading values (≤0.500) and were less correlated with the four described determinants. From these, three are related to traffic safety and security, and the remaining to sidewalk facilities. Contrary to previous research [48,76] our findings suggest that “walk in streets with low traffic speed” and “walk in streets with ≤2 traffic lanes” were not associated with perceived walkability. In fact, in our previous research [55], traffic safety was not identified as a main barrier preventing utilitarian walking in both cities. Similarly, in a study conducted in Porto, Jabbari et al. [40] also concluded that traffic safety was not a critical factor for pedestrians. The relatively low importance of traffic safety in these cities could be explained by the compact urban layout of both cities, particularly of their city centres, characterised by irregular and narrow streets, which usually only have one or two lanes. The existence of low-speed limits (Zones 30), pedestrian-only streets and temporary pedestrian areas over the weekends could also explain the described perceptions about traffic safety. In turn, the low loading of the public security attribute (“walk in streets with many pedestrians”) is in line with other studies confirming that crime, incivilities, and urban violence are not perceived as a problem deterring people from walking in many European cities [5].
The remaining attributes with low loadings were “walk on sidewalks with street furniture”, “walk on sidewalks with low slopes”, and “walk on sidewalks with trees/greenery”. Slopes and the presence of trees are attributes particularly known for their influence on creating comfortable and attractive pedestrian environments [4,76]. Thus, except for some areas in the historic centre of Porto near the Douro River, both cities are relatively flat, which could explain the low correlation of slopes with walking. The low loading of the attribute “walk on sidewalks with trees/greenery” also contradicts previous research [77]. The compact urban structure of both cities and more specifically the large network of porticoes in Bologna, which replace the role performed by vegetation in terms of shade and protection from adverse weather conditions, could explain the low correlation of this attribute with perceived walkability.
The findings from this study have the potential to guide future research and support planning policies. Urban design attributes have often been assessed by street network characteristics (connectivity) at the neighbourhood scale. Our findings suggest that micro design attributes, such as enclosure, transparency, and architectural and landscape diversity, in conjunction with land use mix and residential density, are equally important to create attractive urban ambiances for pedestrians. It also shows that pedestrian infrastructure attributes, namely sidewalk width, sidewalk obstruction, and sidewalk condition, should be included in walkability indexes, since they have been seldom considered in walkability studies and indexes [4,5]. Cities should also be planned to be connected through a network of pedestrian infrastructure and to provide increased proximity to community services. This will make walking trips more convenient and attractive. Finally, walking to public transport requires increasing the density of public transport stops/stations, to reduce walking distances and make public transport more competitive than car trips.
This study has some limitations that should be highlighted. Firstly, the described results are based on self-reported preferences that may contain inconsistencies between reported preferences and individual behaviours. Moreover, the preferences of some demographic groups, especially elderly people, are under-represented. Online samples are regarded as biased because aged people with low levels of education and fewer tech-savvy skills are usually more difficult to target [78]. Secondly, the described evaluation is based on perceived walkability in Bologna and Porto. The replication of these findings should be done with caution. These results could be representative of medium-sized European cities, with compact urban structures, historic urban fabric, and similar socio-demographic and cultural characteristics, but could be less appropriate to describe walkability in cities with other morphologies and habits. For example, the low association of safety and security with perceived walkability could result from specificities particularly in the central areas of these cities. Thirdly, besides fulfilling the minimum cumulative proportion of variance threshold of at least 50%, the share of total variance not explained should be further analysed to identify what those are attributes. This will increase the fit of the model and enable a better understanding of the factors influencing the perceptions of walkability. Fourthly, the questionnaire should desirably be conducted at the same period of time in both cities. Due to administrative reasons, in Bologna the questionnaire was administered during the Spring and Summer, while in Porto it was conducted in the Autumn. The evaluation of the various attributes may also reflect the influence of the natural conditions associated to these seasons on perceived walkability. As shown in some studies [79], the season of the year and weather, directly affect perceptions of public spaces. Fifthly, the findings described in this study do not take into consideration the long-term changes in travel behaviour caused by the COVID-19 pandemic. In particular, attributes such as “walk in areas in close proximity to public transport stops”, “walk in areas in close proximity to car parking”, “walk in streets with wide sidewalks”, and “walk in streets with many pedestrians” could be differently evaluated due to the pandemic effects (maintain social distancing outdoors, greater preference for private and active modes of transport over public transport).
Finally, the described results explain the influence of the various attributes in a single way, without exploring the influence that individual characteristics (age, gender, education), walking purposes (utilitarian/recreational), and geographic context (Bologna/Porto) may have in such evaluation. We intend to report the influence of such variables in future studies.

6. Conclusions

During the last century, cities were planned for cars and the normal response to mobility issues was to increase road capacity [80]. Over the last decades, a policy interest for creating more sustainable, healthier, and inclusive transport systems, including more walkable environments, increased worldwide. Planning cities for people and for pedestrians is not a relatively new concept among local planners, but the relevance of walking as a mode of transport is not yet entirely recognised. This happens for various interconnected main reasons. The historically automobile-oriented development pattern resulted in prioritizing the automobile over the other modes of transport. For this reason, assigning existing lanes for the exclusive use of other modes has been a difficult task [80]. Urban sprawl leads to an increase in commuting distance, which in turn leads to increased car use, making walking less convenient. The lack of walking strategies, political commitment, and resources to ensure the sustained and consistent implementation of pedestrian policies has been another noted reason [81]. Walking is often not adequately measured in transport surveys, resulting in an underestimation of its importance as a mode of transport [82]. As what is not measured is not seen as important, this contributes to neglecting the pedestrian mode in planning practices. And finally, due to the lack of understanding on pedestrians needs and perceptions and on the influence exerted by built environment and streetscape attributes on travel behaviour [83]. To create more pedestrian-friendly cities, planners and decision-makers need to face and overcome these challenges and gaps between theory (what is known) and practice (what has been done).
With this in mind, this study was able to provide some insights in terms of overcoming part of these challenges, namely by shedding light on how the urban space influences citizens’ preferences and experiences related to walking. More specifically, this study examined the perceptions of 1438 individuals from Bologna and Porto regarding the influence of 19 built environment and streetscape attributes on walkability. The evaluation conducted with an Exploratory Factor Analysis revealed that 13 attributes were strongly correlated with four urban determinants of walkability: urban ambiance, pedestrian infrastructure, connectivity and community facilities, and access to other modes of transport.
The findings of this study showed interesting insights in terms of future research and planning policies. More specifically, these findings suggest that streetscape design attributes and the quality of pedestrian infrastructure should be taken into consideration when designing pedestrian planning policies and walkability indexes. Our findings also demonstrated that the scope of walkability is broad and relies on multiple interconnected attributes at different scales. These attributes should be analysed in conjunction. Adopting a “silo” approach that overlooks the linkages between the various urban dimensions associated to walkability could result in inefficient policies, inappropriate methods to evaluate walkability, and in potential variability across geographical units. Our findings support the development of policies and methods that combine urban ambiance, pedestrian infrastructure, street connectivity and community facilities, and access to other modes of transport attributes, namely by creating more compact urban structures. Planners and decision-makers should not view walkability in a siloed approach or as a stand-alone goal, but rather as one of the underwriting factors in creating more walkable, livable, and sustainable cities. These findings could be used to improve both the methodological and empirical foundations for evaluating walkability, as well as to contribute to the ongoing discussions on how land-use and transport planning can be guided to improve a more sustainable mobility in our cities.

Author Contributions

Conceptualisation, F.F. and G.P.; methodology, F.F. and G.P.; software, G.P.; validation, G.P., F.F. and S.T.; formal analysis, G.P.; investigation, F.F. and E.C.; resources, P.R. and M.J.; data curation, F.F. and E.C.; writing—original draft preparation, F.F., G.P. and P.R.; writing—review and editing, S.T., E.C. and M.J.; visualisation, F.F.; supervision, S.T. and R.R.; project administration, R.R. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by the JPI Urban Europe, FCT–PT (ENSUF/0004/2016), MIUR-I, FFG-A, and RPF-CY.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data of this study are available on request from the corresponding author [S.T.].

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Bassett, D.R.; Pucher, J.; Buehler, R.; Thompson, D.; Crouter, S. Walking, Cycling, and Obesity Rates in Europe, North America, and Australia. J. Phys. Act. Health 2008, 5, 795–814. [Google Scholar] [CrossRef] [PubMed]
  2. Keall, M.D.; Shaw, C.; Chapman, R.; Howden-Chapman, P. Reductions in carbon dioxide emissions from an intervention to promote cycling and walking: A case study from New Zealand. Transp. Res. Part D Transp. Environ. 2018, 65, 687–696. [Google Scholar] [CrossRef]
  3. Ribeiro, P.J.; Fonseca, F. Students’ home-university commuting patterns: A shift towards more sustainable modes of transport. Case Stud. Transp. Policy 2022, 10, 954–964. [Google Scholar] [CrossRef]
  4. Arellana, J.; Saltarín, M.; Larrañaga, A.M.; Alvarez, V.; Henao, C.A. Urban walkability considering pedestrians’ perceptions of the built environment: A 10-year review and a case study in a medium-sized city in Latin America. Transp. Rev. 2020, 40, 183–203. [Google Scholar] [CrossRef]
  5. Fonseca, F.; Ribeiro, P.J.G.; Conticelli, E.; Jabbari, M.; Papageorgiou, G.; Tondelli, S.; Ramos, R.A.R. Built environment attributes and their influence on walkability. Int. J. Sustain. Transp. 2021, 16, 660–679. [Google Scholar] [CrossRef]
  6. Habibian, M.; Hosseinzadeh, A. Walkability index across trip purposes. Sustain. Cities Soc. 2018, 42, 216–225. [Google Scholar] [CrossRef]
  7. Forsyth, A. What is a walkable place? The walkability debate in urban design. Urban Des. Int. 2015, 20, 274–292. [Google Scholar] [CrossRef]
  8. Dovey, K.; Pafka, E. What is walkability? The urban DMA. Urban Stud. 2019, 57, 93–108. [Google Scholar] [CrossRef]
  9. Dörrzapf, L.; Kovács-Gyóri, A.; Resch, B.; Zeile, P. Defining and assessing walkability: An concept for an integrated approach using surveys, biosensors and geospatial analysis. Urban Dev. Issues 2019, 62, 5–15. [Google Scholar] [CrossRef] [Green Version]
  10. Cervero, R.; Kockelman, K. Travel demand and the 3Ds: Density, diversity, and design. Transp. Res. Part D Transp. Environ. 1997, 2, 199–219. [Google Scholar] [CrossRef]
  11. Ewing, R.; Cervero, R. Travel and the Built Environment: A meta-analysis. J. Am. Plan. Assoc. 2010, 76, 265–294. [Google Scholar] [CrossRef]
  12. Frank, L.D.; Schmid, T.L.; Sallis, J.F.; Chapman, J.; Saelens, B.E. Linking objectively measured physical activity with objectively measured urban form: Findings from SMARTRAQ. Am. J. Prev. Med. 2005, 28, 117–125. [Google Scholar] [CrossRef] [PubMed]
  13. Frank, L.D.; Sallis, J.F.; Saelens, B.E.; Leary, L.; Cain, K.; Conway, T.L.; Hess, P.M. The development of a walkability index: Application to the Neighborhood Quality of Life Study. Br. J. Sports Med. 2010, 44, 924–933. [Google Scholar] [CrossRef] [PubMed]
  14. Moran, M.R.; Eizenberg, E.; Plaut, P. Getting to Know a Place: Built Environment Walkability and Children’s Spatial Representation of Their Home-School (h–s) Route. Int. J. Environ. Res. Public Health 2017, 14, 607. [Google Scholar] [CrossRef] [Green Version]
  15. Williams, G.C.; Borghese, M.M.; Janssen, I. Neighborhood walkability and objectively measured active transportation among 10–13 year olds. J. Transp. Health 2018, 8, 202–209. [Google Scholar] [CrossRef]
  16. Iroz-Elardo, N.; Adkins, A.; Ingram, M. Measuring perceptions of social environments for walking: A scoping review of walkability surveys. Health Place 2020, 67, 102468. [Google Scholar] [CrossRef]
  17. Koohsari, M.J.; Badland, H.; Sugiyama, T.; Mavoa, S.; Christian, H.; Giles-Corti, B. Mismatch between Perceived and Objectively Measured Land Use Mix and Street Connectivity: Associations with Neighborhood Walking. J. Hered. 2014, 92, 242–252. [Google Scholar] [CrossRef] [Green Version]
  18. Andrews, G.J.; Hall, E.; Evans, B.; Colls, R. Moving beyond walkability: On the potential of health geography. Soc. Sci. Med. 2012, 75, 1925–1932. [Google Scholar] [CrossRef]
  19. Jensen, W.; Stump, T.; Brown, B.; Werner, C.; Smith, K. Walkability, complete streets, and gender: Who benefits most? Health Place 2017, 48, 80–89. [Google Scholar] [CrossRef]
  20. Wallmann, B.; Bucksch, J.; Froboese, I. The association between physical activity and perceived environment in German adults. Eur. J. Public Health 2011, 22, 502–508. [Google Scholar] [CrossRef] [Green Version]
  21. Shashank, A.; Schuurman, N. Unpacking walkability indices and their inherent assumptions. Health Place 2018, 55, 145–154. [Google Scholar] [CrossRef] [PubMed]
  22. Liao, B.; Berg, P.E.V.D.; van Wesemael, P.; Arentze, T.A. Individuals’ perception of walkability: Results of a conjoint experiment using videos of virtual environments. Cities 2022, 125, 103650. [Google Scholar] [CrossRef]
  23. Adkins, A.; Barillas-Longoria, G.; Martínez, D.N.; Ingram, M. Differences in social and physical dimensions of perceived walkability in Mexican American and non-hispanic white walking environments in Tucson, Arizona. J. Transp. Health 2019, 14. [Google Scholar] [CrossRef] [PubMed]
  24. Jun, H.-J.; Hur, M. The relationship between walkability and neighborhood social environment: The importance of physical and perceived walkability. Appl. Geogr. 2015, 62, 115–124. [Google Scholar] [CrossRef]
  25. Perry, C.; Herting, J.; Berke, E.; Nguyen, H.; Vernez, A.; Beresford, S.; Ockene, J.; Manson, J.; LaCroix, A. Does neighborhood walkability moderate the effects of intrapersonal characteristics on amount of walking in post-menopausal women? Health Place 2013, 21, 39–45. [Google Scholar] [CrossRef] [Green Version]
  26. Dörrzapf, L.; Zeile, P.; Brocza, U.; Schwomma, Y.; Resch, B.; Kovács-Győri, A.; Berger, M. New integrated walkability research approach. In Proceedings of the REAL CORP 2019, Karlsruhe, Germany, 2–4 April 2019. [Google Scholar]
  27. Fancello, G.; Congiu, T.; Tsoukiàs, A. Mapping walkability. A subjective value theory approach. Socio-Economic Plan. Sci. 2020, 72, 100923. [Google Scholar] [CrossRef]
  28. Saadi, I.; Aganze, R.; Moeinaddini, M.; Asadi-Shekari, Z.; Cools, M. A Participatory Assessment of Perceived Neighbourhood Walkability in a Small Urban Environment. Sustainability 2021, 14, 206. [Google Scholar] [CrossRef]
  29. Molina-García, J.; Campos, S.; García-Massó, X.; Herrador-Colmenero, M.; Gálvez-Fernández, P.; Molina-Soberanes, D.; Queralt, A.; Chillón, P. Different neighborhood walkability indexes for active commuting to school are necessary for urban and rural children and adolescents. Int. J. Behav. Nutr. Phys. Act. 2020, 17, 1–11. [Google Scholar] [CrossRef]
  30. Saelens, B.E.; Sallis, J.F.; Black, J.B.; Chen, D. Neighborhood-Based Differences in Physical Activity: An Environment Scale Evaluation. Am. J. Public Health 2003, 93, 1552–1558. [Google Scholar] [CrossRef]
  31. Taleai, M.; Amiri, E.T. Spatial multi-criteria and multi-scale evaluation of walkability potential at street segment level: A case study of tehran. Sustain. Cities Soc. 2017, 31, 37–50. [Google Scholar] [CrossRef]
  32. Forsyth, A.; Oakes, J.; Schmitz, K.; Hearst, M. Does residential density increase walking and other physical activity? Urban Stud. 2007, 44, 679–697. [Google Scholar] [CrossRef]
  33. Huang, R.; Moudon, A.; Zhou, C.; Saelens, B. Higher residential and employment densities are associated with more objectively measured walking in the home neighbourhood. J. Transp. Health 2019, 12, 142–151. [Google Scholar] [CrossRef] [PubMed]
  34. Ribeiro, A.I.; Hoffimann, E. Development of a Neighbourhood Walkability Index for Porto Metropolitan Area. How Strongly Is Walkability Associated with Walking for Transport? Int. J. Environ. Res. Public Health 2018, 15, 2767. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  35. Clark, A.; Scott, D.M.; Yiannakoulias, N. Examining the relationship between active travel, weather, and the built environment: A multilevel approach using a GPS-enhanced dataset. Transportation 2013, 41, 325–338. [Google Scholar] [CrossRef] [Green Version]
  36. Shammas, T.; Escobar, F. Comfort and time-based walkability index design: A GIS-based proposal. Int. J. Environ. Res. Public Health 2019, 16, 2850. [Google Scholar] [CrossRef] [Green Version]
  37. Wang, C.; Chen, N.; Tian, G. Do accessibility and clustering affect active travel behavior in Salt Lake City? Transp. Res. Part D 2021, 90, 1026552021. [Google Scholar] [CrossRef]
  38. Vale, D. Does commuting time tolerance impede sustainable urban mobility? Analysing the impacts on commuting behaviour as a result of workplace relocation to a mixed-use centre in Lisbon. J. Transp. Geogr. 2013, 32, 38–48. [Google Scholar] [CrossRef]
  39. Moreno, C.; Allam, Z.; Chabaud, D.; Gall, C.; Pratlong, F. Introducing the “15-Minute City”: Sustainability, Resilience and Place Identity in Future Post-Pandemic Cities. Smart Cities 2021, 4, 93–111. [Google Scholar] [CrossRef]
  40. Jabbari, M.; Fonseca, F.; Ramos, R. Combining multi-criteria and space syntax analysis to assess a pedestrian network: The case of Oporto. J. Urban Des. 2017, 23, 23–41. [Google Scholar] [CrossRef]
  41. Boulange, C.; Pettit, C.; Gunn, L.; Giles-Corti, B.; Badland, H. Improving planning analysis and decision making: The development and application of a Walkability Planning Support System. J. Transp. Geogr. 2018, 69, 129–137. [Google Scholar] [CrossRef]
  42. Cruise, S.; Hunter, R.F.; Kee, F.; Donnelly, M.; Ellis, G.; Tully, M.A. A comparison of road- and footpath-based walkability indices and their associations with active travel. J. Transp. Health 2017, 6, 119–127. [Google Scholar] [CrossRef] [Green Version]
  43. Ferrer, S.; Ruiz, T.; Mars, L. A qualitative study on the role of the built environment for short walking trips. Transp. Res. Part F: Traffic Psychol. Behav. 2015, 33, 141–160. [Google Scholar] [CrossRef]
  44. Herrmann-Lunecke, M.G.; Mora, R.; Vejares, P. Perception of the built environment and walking in pericentral neighbourhoods in Santiago, Chile. Travel Behav. Soc. 2021, 23, 192–206. [Google Scholar] [CrossRef]
  45. Taleai, M.; Yameqani, A.S. Integration of GIS, remote sensing and Multi-Criteria Evaluation tools in the search for healthy walking paths. KSCE J. Civ. Eng. 2017, 22, 279–291. [Google Scholar] [CrossRef]
  46. Gaglione, F.; Cottrill, C.; Gargiulo, C. Urban services, pedestrian networks and behaviors to measure elderly accessibility. Transp. Res. Part D Transp. Environ. 2020, 90, 102687. [Google Scholar] [CrossRef]
  47. Ruiz-Padillo, A.; Pasqual, F.M.; Uriarte, A.M.L.; Cybis, H.B.B. Application of multi-criteria decision analysis methods for assessing walkability: A case study in Porto Alegre, Brazil. Transp. Res. Part D Transp. Environ. 2018, 63, 855–871. [Google Scholar] [CrossRef]
  48. Giles-Corti, B.; Wood, G.; Pikora, T.; Learnihan, V.; Bulsara, M.; Van Niel, K.; Timperio, A.; McCormack, G.; Villanueva, K. School site and the potential to walk to school: The impact of street connectivity and traffic exposure in school neighborhoods. Health Place 2011, 17, 545–550. [Google Scholar] [CrossRef]
  49. Lovasi, G.; Jacobson, J.; Quinn, J.; Neckerman, K.; Ashby-Thompson, M.; Rundle, A. Is the environment near home and school associated with physical activity and adiposity of urban preschool children? J. Urban Health 2011, 88, 1143–1157. [Google Scholar] [CrossRef] [Green Version]
  50. Zhang, C.; Zhou, B.; Chen, G.; Chen, F. Quantitative analysis of pedestrian safety at uncontrolled multi-lane mid-block crosswalks in China. Accid. Anal. Prev. 2017, 108, 19–26. [Google Scholar] [CrossRef]
  51. Foster, S.; Wood, L.; Christian, H.; Knuiman, M.; Giles-Corti, B. Planning safer suburbs: Do changes in the built environment influence residents’ perceptions of crime risk? Soc. Sci. Med. 2013, 97, 87–94. [Google Scholar] [CrossRef]
  52. Ewing, R.; Handy, S. Measuring the Unmeasurable: Urban Design Qualities Related to Walkability. J. Urban Des. 2009, 14, 65–84. [Google Scholar] [CrossRef]
  53. Yin, L. Street level urban design qualities for walkability: Combining 2D and 3D GIS measures. Comput. Environ. Urban Syst. 2017, 64, 288–296. [Google Scholar] [CrossRef]
  54. Nichani, V.; Vena, J.E.; Friedenreich, C.M.; Christie, C.; McCormack, G.R. A population-based study of the associations between neighbourhood walkability and different types of physical activity in Canadian men and women. Prev. Med. 2019, 129, 105864. [Google Scholar] [CrossRef] [PubMed]
  55. Fonseca, F.; Conticelli, E.; Papageorgiou, G.; Ribeiro, P.; Jabbari, M.; Tondelli, S.; Ramos, R. Levels and characteristics of utilitarian walking in the cities of Bologna and Porto. Sustainability 2021, 13, 3064. [Google Scholar] [CrossRef]
  56. Fonseca, F.; Conticelli, E.; Papageorgiou, G.; Ribeiro, P.; Jabbari, M.; Tondelli, S.; Ramos, R. Use and Perceptions of Pedestrian Navigation Apps: Findings from Bologna and Porto. ISPRS Int. J. Geo-Inform. 2021, 10, 446. [Google Scholar] [CrossRef]
  57. Jam, A.S.; Tabatabaei, M.R.; Mosaffaie, J.; Soltani, M.J.; Shadmani, A. Barriers to participatory implementation of soil conservation projects: Perspectives and priorities. Environ. Sci. Policy 2022, 131, 36–45. [Google Scholar] [CrossRef]
  58. ISTAT-Istituto Nazionale di Statistica. Annual Estimated Population. 2019. Available online: www.istat.it (accessed on 1 February 2022).
  59. SP-Statistics Portugal. Annual Estimated Population. 2019. Available online: www.ine.pt (accessed on 1 February 2022).
  60. Herian, M.; Tomkins, A. Citizen satisfaction survey data: A mode comparison of the derived importance-perfomance approach. Am. Rev. Public Adm. 2012, 42, 66–86. [Google Scholar] [CrossRef]
  61. Agyeman, S.; Kwarteng, R.A.; Zurkalnaine, S. Principal component analysis of driver challenges in the shared taxi market in Ghana. Case Stud. Transp. Policy 2018, 7, 73–86. [Google Scholar] [CrossRef]
  62. Martínez, O.; Ramírez, A. Walkability and the built environment: Validation of the Neighborhood Environment Walkability Scale (NEWS) for urban areas in Mexico. Qual. Quant. 2018, 52, 703–718. [Google Scholar] [CrossRef]
  63. Shrestha, N. Factor Analysis as a Tool for Survey Analysis. Am. J. Appl. Math. Stat. 2021, 9, 4–11. [Google Scholar] [CrossRef]
  64. Zavareh, M.; Mehdizadeh, M.; Nordfjærn, T. “If I know when you will arrive, I will let you walk to school:” The role of information technology. J. Saf. Res. 2020, 72, 267–277. [Google Scholar] [CrossRef] [PubMed]
  65. Mohanty, R.N.; Chani, P.S.; Mohanta, A. Measuring the impact of the built environment on pedestrians in the old Bhubaneswar precinct. J. Heritage Tour. 2020, 16, 181–200. [Google Scholar] [CrossRef]
  66. Lindelöw, D.; Svensson, Å.; Brundell-Freij, K.; Hiselius, L.W. Satisfaction or compensation? The interaction between walking preferences and neighbourhood design. Transp. Res. Part D Transp. Environ. 2017, 50, 520–532. [Google Scholar] [CrossRef]
  67. Kissi, E.; Boateng, E.B.; Adjei-Kumi, T.; Badu, E. Principal component analysis of challenges facing the implementation of value engineering in public projects in developing countries. Int. J. Constr. Manag. 2016, 17, 142–150. [Google Scholar] [CrossRef]
  68. Thibaud, J.-P. The Sensory Fabric of Urban Ambiances. Senses Soc. 2011, 6, 203–215. [Google Scholar] [CrossRef] [Green Version]
  69. Ramezani, S.; Laatikainen, T.; Hasanzadeh, K.; Kyttä, M. Shopping trip mode choice of older adults: An application of activity space and hybrid choice models in understanding the effects of built environment and personal goals. Transportation 2019, 48, 505–536. [Google Scholar] [CrossRef] [Green Version]
  70. Yin, L.; Wang, Z. Measuring visual enclosure for street walkability: Using machine learning algorithms and Google Street View imagery. Appl. Geogr. 2016, 76, 147–153. [Google Scholar] [CrossRef]
  71. Bahrainy, H.; Khosravi, H. The impact of urban design features and qualities on walkability and health in under-construction environments: The case of Hashtgerd New Town in Iran. Cities 2013, 31, 17–28. [Google Scholar] [CrossRef]
  72. Vural-Arslan, T.; Dostolu, N.; Köprülü-Babanci, Ö.; Aknctürk, N. Sustainable revitalisation as a tool for regenerating the attractiveness of an inner-city historic commercial district: Han district as a case. Urban Des. Int. 2011, 16, 188–201. [Google Scholar] [CrossRef]
  73. Corazza, M.V.; Di Mascio, P.; Moretti, L. Managing sidewalk pavement maintenance: A case study to increase pedestrian safety. J. Traffic Transp. Eng. English Ed. 2016, 3, 203–214. [Google Scholar] [CrossRef] [Green Version]
  74. ISTAT-Istituto Nazionale di Statistica. 15th Population and Housing Census 2011. Available online: www.istat.it (accessed on 11 March 2022).
  75. SP-Statistics Portugal. Census 2011 Data. Available online: www.ine.pt (accessed on 11 March 2022).
  76. Golan, Y.; Wilkinson, N.; Henderson, J.M.; Weverka, A. Gendered walkability: Building a daytime walkability index for women. J. Transp. Land Use 2019, 12. [Google Scholar] [CrossRef]
  77. Kweon, B.-S.; Rosenblatt-Naderi, J.; Ellis, C.; Shin, W.-H.; Danies, B. The Effects of Pedestrian Environments on Walking Behaviors and Perception of Pedestrian Safety. Sustainability 2021, 13, 8728. [Google Scholar] [CrossRef]
  78. Saloniki, E.-C.; Malley, J.; Burge, P.; Lu, H.; Batchelder, L.; Linnosmaa, I.; Trukeschitz, B.; Forder, J. Comparing internet and face-to-face surveys as methods for eliciting preferences for social care-related quality of life: Evidence from England using the ASCOT service user measure. Qual. Life Res. 2019, 28, 2207–2220. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  79. Szczepańska, A.; Pietrzyk, K. Seasons of the Year and Perceptions of Public Spaces. Landsc. J. 2021, 40, 19–35. [Google Scholar] [CrossRef]
  80. Lindau, L.A.; Hidalgo, D.; de Almeida Lobo, A. Barriers to planning and implementing Bus Rapid Transit systems. Res. Transp. Econ. 2014, 48, 9–15. [Google Scholar] [CrossRef]
  81. McClintock, H. Practitioners’ Take-up of Professional Guidance and Research Findings: Planning for Cycling and Walking in the UK. Plan. Pr. Res. 2001, 16, 193–203. [Google Scholar] [CrossRef]
  82. Sauter, D.; Pharoah, T.; Tight, M.; Martinson, R.; Wedderburn, M. International walking data standard, treatment of walking in travel surveys internationally standardized monitoring methods of walking and public space. Walk 21. Available online: http://www.measuring-walking.org/ (accessed on 3 May 2022).
  83. Mateo-Babiano, I. Pedestrian’s needs matters: Examining Manila’s walking environment. Transp. Policy 2016, 45, 107–115. [Google Scholar] [CrossRef]
Figure 1. Location of Bologna and Porto.
Figure 1. Location of Bologna and Porto.
Sustainability 14 09089 g001
Figure 2. Scree plot model for exploratory factor analysis.
Figure 2. Scree plot model for exploratory factor analysis.
Sustainability 14 09089 g002
Figure 3. Urban determinants of walkability and main factors.
Figure 3. Urban determinants of walkability and main factors.
Sustainability 14 09089 g003
Table 1. Built environment and streetscape attributes included in the questionnaire.
Table 1. Built environment and streetscape attributes included in the questionnaire.
Type of AttributeDescription
Built environment1. Walk in areas in close proximity to public transport stops
Built environment2. Walk in areas in close proximity to car parking
Built environment3. Walk in areas with high street connectivity
Built environment4. Walk in areas in close proximity to community facilities
Streetscape5. Walk in streets with low traffic speed
Streetscape6. Walk in streets with ≤2 traffic lanes
Streetscape7. Walk in streets with wide sidewalks
Streetscape8. Walk on sidewalks in good condition
Streetscape9. Walk on unobstructed sidewalks
Streetscape10. Walk on sidewalks with street furniture
Streetscape11. Walk on sidewalks with low slopes
Streetscape12. Walk on sidewalks with trees/greenery
Streetscape13. Walk in streets with many pedestrians
Built environment14. Walk in shopping streets/areas
Built environment15. Walk in areas with high residential density
Built environment16. Walk in areas with mixed land uses
Streetscape17. Walk in streets providing enclosure
Streetscape18. Walk in streets with architectural and landscape diversity
Streetscape19. Walk in streets providing transparency
Table 2. Sample description.
Table 2. Sample description.
VariableAttributesQuestionnairePopulation 2019
BolognaPortoBolognaPorto
Total%Total%Total%Total%
GenderFemale50758.634159.5206,58952.7119,22855.0
Male35841.423240.5185,39547.397,37845.0
Age≤24 years old849.711019.278,41020.047,84622.1
25–44 years old26630.823641.2103,97326.546,82121.6
45–64 years old47755.121437.3112,55428.760,22327.8
≥65 years old384.4.132.397,04724.861,71628.5
EducationUndergraduates56264.930853.8308,81678.8163,62175.5
Graduates30335.126546.283,16821.252,98524.5
OccupationStudent11112.815527.051,05415.642,08920.9
Employed73585.040270.2165,76850.588,45243.8
Unemployed/retired192.2162.8111,41433.971,23535.3
Type of
pedestrian
Resident48055.537765.8391,984100.0216,606100.0
Commuter36241.816428.6----
Tourist/visitor232.7325.6----
Source (Population data): ISTAT [58]; SP [59].
Table 3. Descriptive results of the evaluation made by the participants (n = 1438).
Table 3. Descriptive results of the evaluation made by the participants (n = 1438).
AttributesMeanMedianSD
8. Walk on sidewalks in good condition4.475.000.83
9. Walk on unobstructed sidewalks4.375.000.90
7. Walk in streets with wide sidewalks4.224.000.95
12. Walk on sidewalks with trees/greenery4.144.000.93
10. Walk on sidewalks with street furniture4.084.000.99
4. Walk in areas closer to community facilities4.014.001.06
3. Walk in areas with high street connectivity3.954.001.10
5. Walk in streets with low traffic speed3.784.001.20
18. Walk in streets with architectural and landscape diversity3.574.001.06
1. Walk in areas closer to public transport stops3.514.001.33
13. Walk in streets with many pedestrians3.393.001.10
16. Walk in areas with mixed land uses3.213.001.06
11. Walk on sidewalks with low slopes3.193.001.24
14. Walk in shopping streets/areas3.043.001.19
19. Walk in streets providing transparency2.943.001.09
6. Walk in streets with ≤2 traffic lanes2.933.001.23
2. Walk in areas closer to car parking2.883.001.44
15. Walk in areas with high residential density2.673.001.08
17. Walk in streets providing enclosure2.402.001.06
Table 4. Results of the KMO and Bartlett’s Test.
Table 4. Results of the KMO and Bartlett’s Test.
Kaiser–Meyer–Olkin Measure of Sampling Adequacy0.902
Bartlett’s Test of SphericityApprox. Chi-Square10,462.1
df171
Sig.0.000
Table 5. Communalities for the 19 built environment and streetscape attributes.
Table 5. Communalities for the 19 built environment and streetscape attributes.
Built Environment and Streetscape AttributesInitialExtraction
15. Walk in areas with high residential density1.0000.659
1. Walk in areas closer to public transport stops1.0000.647
4. Walk in areas closer to public facilities1.0000.643
3. Walk in areas with high street connectivity1.0000.631
8. Walk on sidewalks in good condition1.0000.623
14. Walk in shopping streets/areas1.0000.616
16. Walk in areas with mixed land uses1.0000.616
19. Walk in streets providing transparency1.0000.577
2. Walk in areas closer to car parking1.0000.575
9. Walk on unobstructed sidewalks1.0000.574
7. Walk in streets with wide sidewalks1.0000.570
17. Walk in streets providing enclosure1.0000.569
18. Walk in streets with architectural and landscape diversity1.0000.548
13. Walk in streets with many pedestrians1.0000.493
5. Walk in streets with low traffic speed1.0000.469
10. Walk on sidewalks with street furniture1.0000.466
11. Walk on sidewalks with low slopes1.0000.458
12. Walk on sidewalks with trees/greenery1.0000.386
6. Walk in streets with ≤2 traffic lanes1.0000.247
Table 6. Eigenvalues and total variance explained.
Table 6. Eigenvalues and total variance explained.
ComponentTotalInitial EigenvaluesExtraction Sums of Squared LoadingsRotation Sums of Squared Loadings
Variance
%
Cumulative
%
TotalVariance
%
Cumulative
%
TotalVariance
%
Cumulative
%
C16.28533.07833.0786.28533.07833.0784.15921.88921.889
C22.33912.31245.3902.33912.31245.3903.19316.80738.696
C31.4607.68353.0721.4607.68353.0721.97710.40349.099
C41.0575.56158.6341.0575.56158.6341.8129.53558.634
C50.8894.67863.311
C60.8074.24867.560
C70.7523.95971.519
C80.5753.02774.546
C90.5542.91877.464
C100.5482.88780.351
C110.5192.73283.083
C120.4902.57785.660
C130.4702.47688.136
C140.4472.35190.487
C150.4242.23492.721
C160.4072.14294.863
C170.3671.93096.793
C180.3381.77998.572
C190.2711.428100.000
Table 7. Rotated Component Matrix.
Table 7. Rotated Component Matrix.
Factors and AttributesC1C2C3C4
Urban Ambiance
15. Walk in areas with high residential density0.823
17. Walk in streets providing enclosure0.794
14. Walk in shopping streets/areas0.743
16. Walk in areas with mixed land uses0.729
19. Walk in streets providing transparency0.676
18. Walk in streets with architectural and landscape diversity0.557
Pedestrian infrastructure
8. Walk on sidewalks in good condition 0.795
9. Walk on unobstructed sidewalks 0.770
7. Walk on streets with wide sidewalks 0.730
Connectivity and community facilities
4. Walk in areas closer to community facilities 0.717
3. Walk in areas with high street connectivity 0.705
Access to other modes of transport
2. Walk in areas closer to car parking 0.806
1. Walk in areas closer to public transport stops 0.758
Cronbach’s alpha (α)0.850.790.700.68
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Fonseca, F.; Papageorgiou, G.; Tondelli, S.; Ribeiro, P.; Conticelli, E.; Jabbari, M.; Ramos, R. Perceived Walkability and Respective Urban Determinants: Insights from Bologna and Porto. Sustainability 2022, 14, 9089. https://doi.org/10.3390/su14159089

AMA Style

Fonseca F, Papageorgiou G, Tondelli S, Ribeiro P, Conticelli E, Jabbari M, Ramos R. Perceived Walkability and Respective Urban Determinants: Insights from Bologna and Porto. Sustainability. 2022; 14(15):9089. https://doi.org/10.3390/su14159089

Chicago/Turabian Style

Fonseca, Fernando, George Papageorgiou, Simona Tondelli, Paulo Ribeiro, Elisa Conticelli, Mona Jabbari, and Rui Ramos. 2022. "Perceived Walkability and Respective Urban Determinants: Insights from Bologna and Porto" Sustainability 14, no. 15: 9089. https://doi.org/10.3390/su14159089

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop