Unfolding time,race and class inequalities to access leisure

Abstract

The access to leisure activities is an important element to understand the potential participation and integration of individuals in the society. Despite its importance, urban planners in large urban centers in developing countries seek to prioritize access to mandatory activities. This study quantifies the accessibility to leisure and its inequalities in the municipality of São Paulo, considering the opening hours of leisure opportunities and racial and class population groups. Tracking data from buses and TomTom speed profile were used in the public and the private transport networks, respectively, to analyze and compare accessibility to parks and cultural equipment. A multitemporal analysis was performed to better understand the fluctuation of accessibility to leisure through different hours considering the opening hours of parks and cultural equipment. The population was stratified into four groups according to race and class (higher black, higher white, lower black, and lower white) to perform accessibility inequalities analysis. Results show that accessibility to leisure is higher for private transport users, it decreases from the central to the peripheral areas, and it changes significantly during the day due to traffic conditions, transit supply, and leisure opportunities opening hours. The Lorenz curves, Gini, and the Palma coefficients showed a highly unequal level of accessibility to leisure for different population groups, with the low-black population having the lowest level of leisure accessibility. Our findings may support policy makers in designing strategies to provide more spatial equity in the access to leisure opportunities.

Keywords

Accessibility to leisure accessibility inequalities vehicle tracking data private transport public transport

Introduction

The access to leisure is an important element to understand the potential participation and integration of individuals in the society. The benefits of leisure activities, including the access to green areas, is well known and includes helping individuals recover from the physical and mental stress of everyday urban life (Grahn and Stigsdotter, 2010), improving the degree of social interaction (Sullivan et al., 2004), and to have a better health in general (De Vries et al., 2003). Despite its importance, transport planners in developing countries typically focus on transport infrastructure for mandatory trips, such as work and education (Vecchio et al., 2020). One straightforward way to identify the “benefits provided by a transportation/land-use system” (Ben-Akiva and Lerman, 1979) is through accessibility measures.

The accessibility concept has become central to urban planning (Batty, 2009), and is defined by Hansen (1959) as “the potential of opportunities for interaction.” As the joint result of a transportation network and the geographical distribution of activities (Paéz et al., 2012), accessibility is the key to understanding the distributional effects of transport and land-use policies (Van Wee and Geurs, 2011). These distributional effects and who benefits most from accessibility are related to the location of public transport infrastructure, spatial patterns of land use, and the location of different population groups (Foth et al., 2013).

Although urban planners are concerned with the equitable distribution of public amenities (Tahmasbi et al., 2019), residents of a city experience different levels of accessibility due to the central and peripheral nature of cities (Martens, 2012), and transportation inefficiencies caused by urban sprawl and population growth (Mohri et al., 2021). Accessibility inequality can also be partly explained by the preferences of the privileged wealthy population, which takes most of the transportation benefits, while the unprivileged low-socioeconomic class population is more exposed to the transportation externalities (Bills and Walker, 2017). In Latin American cities, where the accessibility gap between the wealthy and the poor population is severe, the small wealthy population have access to diverse and good services, while the vast majority of the population has insufficient levels of accessibility to essential services (Hernandez, 2018).

In this context, this paper aims to calculate the accessibility to leisure in a large Latin American urban center (São Paulo, Brazil), and to unveil its inequalities considering different class and race population groups. To do so, tracking data for both the public and the private transport modes were used to create transport networks in a GIS environment. Two inequality coefficients were used to indicate inequalities in accessibility to leisure for the population groups.

The remainder of this article is organized as follows. In Section 2, accessibility to leisure and accessibility inequalities is reviewed. In Section 3, the methodology is presented. In Section 4, the results of the accessibility to leisure and its inequalities are presented. Discussions and conclusions are presented in Section 5.

Accessibility to leisure and inequalities

Accessibility to leisure and its inequalities has been the subject of study of a few researchers, with a special focus on parks and green areas (Siqueira-Gay et al., 2019; Wang and Lan, 2019; Wei, 2017; Xing et al., 2018). The distribution of leisure opportunities has become an important issue in spatial planning with the growing concern of urban population with quality of life (De Vries et al., 2003). As with other types of opportunities, access to leisure can be unevenly distributed. Studies have identified that inequalities in accessibility to leisure may be related to population characteristics, such as income (Hoffimann et al., 2017; Wang et al., 2015; Yasumoto et al., 2014), ethnic and religious groups (Comber et al., 2008; Omer, 2006), age (Gupta et al., 2016), and gender (Rojas et al., 2016).

The income characteristic of population and its relationship with accessibility to parks is explored by Yasumoto et al. (2014) in Yokohama, Japan. Results show that accessibility to parks is positively associated with land prices in Yokohama, and that new parks are in more affluent communities, and yet encourage further move-in of affluent population (Yasumoto et al., 2014). In a similar line, the study of Wang et al., (2015) indicated that lower socioeconomic status neighborhoods have a lower perceived park accessibility in Brisbane, Australia. Furthermore, the study of Hoffiman et al. (2017) in Porto, Portugal, found that mean distance to green space increases with neighborhood deprivation. Moreover, green spaces in the more deprived neighborhoods lack safety and equipment to engage in leisure activities, and present signs of damage (Hoffiman et al., 2017).

Regarding inequalities in access to leisure related to the ethnic composition of the population, Omer (2006) found that in Yaffo, Israel, while 92.5% of the Jewish population has access to parks, this percentage drops to only 77.9% of the Arab population. In Leicester, England, a study by Comber et al. (2008) found that the Indian, Hindu and Sikh groups have limited access to green space, when compared to Christian, Jewish, Muslim, and other religious and non-religious groups.

The age-related accessibility to urban green spaces is explored by Gupta et al. (2016) in Delhi, India. The authors found that accessibility is particularly low for parks designed for early age children. In the case of the cities of Valdivia and Temuco, Chile, the study by Rojas et al. (2016) found that variations in accessibility to parks tend to be directed by age and gender, and less by income.

Regarding the accessibility inequalities to cultural equipment, Yang et al. (2018) explored walking accessibility to cultural centers in Xiamei, China. The authors found that walking accessibility to cultural centers has a positive impact on housing prices, which can lead to accessibility inequalities related to income. In a similar line, Siqueira-Gay et al. (2019) investigated spatial inequalities in accessibility to several opportunities, including cultural centers, in São Paulo, Brazil. Results indicated that the low-income population has low access to cultural centers.

The contribution of this paper includes the calculation of accessibility to parks and cultural equipment in São Paulo, Brazil, which is a region with a lack of studies related to accessibility, especially to non-mandatory opportunities. In this paper, the use of multitemporal tracking data for both public and private transport modes and the inclusion of opening hours for parks and cultural equipment allows the calculation of accessibility during each hour of three different days (Saturday, Sunday, and Wednesday). In addition, the stratification of the population according to their class (high and low) and race (black and white), and the use of Lorenz curves, Gini, and Palma inequality coefficients allowed an overview of the inequalities of accessibility to parks and cultural equipment in São Paulo.

Study design and methods

This section presents the study design including a brief description of the study area, data, and metrics used to calculate the accessibility to leisure (parks and cultural equipment) and the inequality coefficients.

Study area

The study area is the municipality of São Paulo, Brazil. São Paulo is the most populated city in South America, with an estimated population of 12.3 million inhabitants¹ and, like most Latin American urban centers, is characterized by deep inequalities between the rich and the poor population (Slovic et al., 2019). These inequalities especially affect how different population groups must face transport barriers to access parks and cultural equipment, which are key places where urban social activities take place.

The distribution of parks, cultural equipment (library, cultural center, museum, art gallery, cinema, theater, and concert hall), income, and public transport are shown in Figure 1. According to Figure 1, parks are apparently well spatially distributed, while cultural equipment’s are concentrated in the city center. The income map shows a typical core-periphery distribution of the population where the low-income population lives in the periphery and the high-income population in the city center. Regarding public transport, it can be seen that most of the supply of mass rail transport system is concentrated in the city center, while the bus system is more distributed.

Figure 1.

Parks, cultural equipment, income, and public transport distribution in São Paulo.

Datasets

The datasets used in this work were the location of parks and cultural equipment, the 2017 origin and destination survey, public and private transport data, and the census data (Table S1 in the Supplementary Material).

The park and cultural equipment data were provided by Geosampa. The park data contains the polygons of all 108 parks of São Paulo. These data were associated with the opening hours² obtained from the local government to better understand the variation of accessibility during the day. The size of the parks and the availability of facilities within it were not considered in this study. The cultural equipment data contains the location points of libraries, cultural centers, museums, art galleries, cinemas, theaters, and concert halls. They were separated into two categories of cultural equipment, the daytime, and the nighttime equipment. The daytime equipment is composed of libraries, cultural centers, museums, cinema, and art galleries, while the nighttime equipment is composed of cinemas, theaters, and concert halls. Cinema is considered both daytime and nighttime equipment. The daytime equipment is open during business hours (from 8a.m. to 6p.m.), while the nighttime equipment is open from 6p.m. to 2a.m.

The OD zones data was provided by the Metrô company (Companhia do Metropolitano, 2007), and its centroids were used as departure points to calculate the accessibility to parks and cultural equipment. The origin and destination microdata were also provided by the Metrô, and it consists of a table with details of each of the trips reported by the survey respondents. The table is composed of approximately 183,000 observations and 126 variables, including the monthly household income variable, trip motive, OD zone of origin and destination of the trip, and expansion factor of the trip. The microdata was used to calculate the travel time threshold for the accessibility measure and the modal share of each OD zone.

The public transport dataset was provided by the SPTrans and it consists of GTFS files with corrected travel times using the automatic vehicle location (AVL) system. Since the aim of this work is to investigate the accessibility to parks and cultural equipment, to consider the weekend is instrumental, thus a Saturday (April 6), Sunday (April 7), and a working day, Wednesday (April 10), were analyzed. While Saturdays and Sundays have differences in terms of public transport supply and are typical days when most parks and amenities receive more visitors, they also have different public transport supply from weekdays, represented in this study by Wednesday. Thus, we capture the differences in accessibility to leisure due to the different transport supply and traffic conditions for these days. The private transport data consists of speed profiles provided by TomTom for the same date of public transport, to capture accessibility to parks and cultural equipment with the same traffic conditions. The Uber database was used to evaluate whether the accessibility results and their inequalities calculated with the TomTom database are maintained. Details of the data and the results regarding Uber database are presented in the Supplementary Material.

The census data was used to stratify the population according to its class and race. Using the same spatial microsimulation methods performed by [removed for peer blind review], we associate the census and the weighting areas data, and the results were aggregated into OD zones.

Accessibility to parks and cultural equipment calculation

The cumulative accessibility measure (equation (1)) was chosen due to its characteristics of ease operationalization, interpretability, and communicability (Geurs and Van Wee, 2004).

A_{i h}^{m} = \sum_{j} O_{j h} I (c_{i j h}^{m} \leq γ)

(1)

where

A_{i h}^{m}

is the accessibility to parks or cultural equipment of OD zone

i

at hour

h

by mode

m

;

O_{j h}

is the park or cultural equipment

j

at hour

h

;

c_{i j h}^{m}

is the travel time for a trip between OD zone

i

and the park or cultural equipment

j

at hour

h

by mode

m

;

γ

is the travel time threshold; and

I

is a logical statement that receives the value 1 when true and 0 when false.

The accessibility to parks and cultural equipment was calculated separately because they have different purposes. Each one of the parks and cultural equipment have specific opening hours (daytime and nighttime) that are considered in the accessibility calculations. Transport modes and departure times also impact travel times due to public transport supply and traffic conditions.

In the case of the public transport, GTFS files were associated with streets segments provided by the Open Street Maps (OSM) to calculate travel times between the centroids of OD zones and the centroids of parks and cultural equipment using the OpenTripPlanner (OTP) for April 6, 7, and 10, 2019. Travel times were calculated hourly to capture variations in accessibility throughout the day.

For private transport, the same dates and times were analyzed using the speed profiles provided by TomTom. The TomTom speed profiles consist of historical speeds patterns for streets segments, which in the case of this study are aggregated by hour. The private transport network was created in a GIS environment by associating the average speeds with a shapefile of street segments provided by TomTom.

Finally, a travel time threshold must be selected to calculate cumulative accessibility. The selection of a travel time threshold is critical because it can highly influence accessibility results and inequalities between population groups. Pereira (2019) investigated this issue by doing a sensitivity analysis of different travel time thresholds when analyzing the impacts of transport interventions in Rio de Janeiro (Brazil). Since there is no consensus in the literature on which travel time threshold to adopt for study areas or opportunity types, we used the origin-destination survey microdata to estimate an average travel time that was adopted as the travel time threshold in the cumulative accessibility measure. Through the microdata, leisure trips were filtered, and it was found that the average travel time was 31 min, considering both public and private transport. Thus, a 30-min threshold was adopted in the calculations of the accessibility to leisure.

As with other accessibility measures, the cumulative measure has some limitations. These limitations include, but are not limited to, the high sensitivity to the travel time threshold (Dong et al., 2006), the inability to incorporate transport user’s characteristics such as age, disability, and gender, and to assign different weights for opportunities within the travel time threshold (Geurs and Van Wee, 2004). Other limitations are the impossibility to incorporate fare costs and competition metrics.

Accessibility inequality analysis

The accessibility inequality analysis was performed by separating the population into four groups according to race and class (high-blacks, high-whites, low-blacks, and low-whites). We chose these four population groups because we understand that they represent extremes of Brazilian society. While wealthy individuals in high socioeconomic classes are mostly white in the municipality of São Paulo and tend to auto-segregate in the city center (Caldeira, 2001), the black and poorer population tends to be segregated in peripheral and precarious areas as identified in recent studies (Bittencourt et al., 2020; Slovic et al., 2019).

The black and the white population groups were created using the self-declared skin color information collected in the Brazilian census. The black population group is formed by the self-declared black and mixed-race individuals, while the white population group is formed by self-declared white race individuals. The high-class group is formed by individuals with higher or lower professional occupations, with monthly income above 10 minimum wages. The low-class group is formed by individuals who are skilled, semi-skilled, or unskilled manual workers, with monthly income of less than 2 minimum wages.

These four population groups were then compared in terms of accessibility to parks and cultural equipment using Lorenz curves, Gini, and Palma coefficients. Three scenarios were considered for the analysis of inequalities: the first in which it is assumed that the entire population has access to private transport mode (scenario private transport); the second in which the entire population uses public transport (scenario public transport); and the third that considers the modal share of public and private transport in each of the OD zones of the municipality of São Paulo (scenario modal share).

The Lorenz curve (Lorenz, 1905), Gini coefficient (Gini, 1912), and Palma ratio (Palma, 2011) were originally developed to measure the concentration of wealth. However, it has shown applications on the assessment of transport equity (Delbosc and Currie, 2011; Giannotti et al., 2021; Jin et al., 2019; Neutens et al., 2010; Pritchard et al., 2019; Ricciardi et al., 2015; Van Wee and Geurs, 2011). In this case, Lorenz curves are graphical representations of the cumulative percentage of the accessibility across the population, while the Gini coefficient (equation (2)) is a metric that indicates the overall accessibility inequality (Delbosc and Currie, 2011):

G = 1 - Σ_{k = 1}^{n} (X_{k} - X_{k - 1}) (Y_{k} + Y_{k - 1})

(2)

where

X_{k}

is the cumulative proportion of the population, with

X_{0} = 0

and

X_{n} = 1

and

Y_{k}

is the cumulative proportion of the accessibility, with

Y_{0} = 0

and

Y_{n} = 1

The Palma coefficient (equation (3)) indicates the inequality considering the average accessibility of the top 10% of the population divided by the average accessibility of the bottom 40% of the population:

P = \frac{{\bar{A}}_{i D 10}}{{\bar{A}}_{i D 1 - D 4}}

(3)

where

{\bar{A}}_{i D 10}

is the average accessibility to parks or cultural equipment of the top 10% of the population;

{\bar{A}}_{i D 1 - D 4}

is the average accessibility to parks or cultural equipment of the bottom 40% of the population;

The overall process flow is presented in Figure 2.

Figure 2.

Overall process flow.

Results and discussions

This section is divided into two subsections. In the first subsection, we present the results of the accessibility to leisure by public and private transport. In the second subsection, we present the results of accessibility inequalities considering the population stratified according to the class and race.

Accessibility to leisure by public and private transport

The accessibility to parks and cultural equipment were calculated for each one of the OD zones considering hourly departures by public and private transport for 06 (Saturday), 07 (Sunday), and 10 (Wednesday) April 2019. The accessibility to parks and cultural equipment for a few selected hours of Saturday is presented in Figure 3. The results considering the Uber database are presented in Figure S1 in the Supplementary Material.

Figure 3.

Accessibility to parks and cultural equipment during Saturday by public and private transport.

As observed in Figure 3, the areas of highest accessibility to parks and cultural equipment are concentrated in the central OD zones of São Paulo, which is where most leisure opportunities are located. However, accessibility to leisure by public and private transport modes present striking differences. Users of public transport can barely reach 30% of the leisure opportunities, while users of private transport can reach up to 60% of the opportunities. The impacts of opening hours of leisure opportunities on accessibility are also observed, as most parks are open from 8 a.m. to 6 p.m. and cultural equipment from 8 a.m. to 2 a.m. The variations in accessibility to leisure caused by traffic and transport supply can be observed when analyzed at 10 a.m. and at 2 p.m., as the number of opportunities available are the same during this period.

The gap between the average accessibility to leisure by public and private transport on Saturday, Sunday, and Wednesday can be seen in Figure 4. The results considering the Uber database are presented in Figure S2 in the Supplementary Material. While users of private transport can reach up to 50% of parks and cultural equipment, users of public transport can barely reach 5% of opportunities. For both public and private transport modes, accessibility to leisure is higher on Sunday. This may be related to the better traffic conditions, despite the lower supply of public transport.

Figure 4.

Percentage of accessible opportunities in 30 min by public (PuT) and private (PrT) transport on Sunday, Saturday, and Wednesday.

Inequalities in accessibility to parks and cultural equipment

The accessibility inequalities were calculated by first stratifying the population by race and class (Figure S3 in Supplementary Material). The high-black population, although concentrated in the city center and east region of São Paulo, is on average 3% of the total population of OD zones, and it reaches not more than 8% of the total population of specific OD zones. Unlike the high-black population, the high-white population is concentrated in the central/southwest region of São Paulo. Some OD zones of these regions have 76% of its population formed by the high-white population. Regarding the low-class population, both the black and the white population are mostly concentrated in the periphery and deprived central areas of São Paulo. Despite the similarity between the low-class populations, the black population is located even far from the central region than the white population. The percentage of samples from each population group is shown in Table S1 in the Supplementary Material.

The results of accessibility to leisure by population groups show a clear pattern that is repeated for all the days analyzed (Figure 5). The high-white population is the one with the highest accessibility to parks and cultural equipment regardless of the day analyzed and the transport mode. The population with the second highest accessibility to leisure is the high-black population followed by a small difference by the low-white population. The population with the lowest accessibility to leisure is the low-black population. These results emphasize the spatial patterns of where these populations reside in the municipality of São Paulo, which is one of the most unequal urban centers in the world. The difference in accessibility to leisure by public and private transport is also sharp, being three to four times greater by private transport. The distribution of the high-class population and the percentage of residents who own at least one car are in the Supplementary Material (Figures S4 and S5).

Figure 5.

Accessibility to leisure by public and private transport according to class and race.

The Lorenz curves are presented in Figure 6. As observed, in general, the Lorenz curves show that the median accessibility to parks and cultural equipment of the high-white population is substantially greater than the median accessibility of other population groups. The high-white population is followed by the high-black population and the low-white population.

Figure 6.

Lorenz curve of accessibility to leisure by class and race.

As noted in the previous analysis, the high-black population and the low-white population have similar accessibility to leisure, here presented by the median accessibility in the Lorenz curves. The low-black population is the group with the lowest median accessibility to leisure, which was already expected, since this population group lives in the peripheral areas, where transport supply and leisure opportunities are scarce.

The inequalities of accessibility to parks and cultural equipment are minimal when considered the private transport. In this case, assuming a hypothetical scenario where the entire population has access to private transport, inequality between individuals would decrease. This scenario is highly unlikely in the reality due to the costs associated with private transport, which prevents the low-class population from having access to private transport. Another aspect is that the current traffic conditions represented by the TomTom data would change if the entire population had access to private transport, increasing travel times, and consequently reducing accessibility to leisure for private transport users, besides worsening living conditions by producing negative externalities like congestion and air pollution. When considering the hypothetical scenario where the entire population use public transport, inequalities are higher than the scenario with private transport. In this case, the population that lives in OD zones with higher transport supply have greater accessibility to leisure, which is reflected in the Lorenz curves. Finally, the inequalities are maximum when considering the proportion of public and private transport users (modal share) in each one of the OD zones. The modal share reflects in higher inequalities due to disparities in accessibility between users of public transport (mostly low-class and located far from leisure opportunities) and private transport (mostly high-class and located closer to leisure opportunities).

The Gini coefficients and Palma are presented in Table 1.

Table 1.

Results of the inequality coefficients.

	Coefficient	Mode	Sat	Sun	Wed
Parks	Gini	Modal share	0.65	0.66	0.65
		Public transport	0.38	0.36	0.38
		Private transport	0.17	0.16	0.17
	Palma	Modal share	38.80	36.45	38.69
		Public transport	6.22	5.47	6.59
		Private transport	2.13	1.99	2.11
Cultural equipment	Gini	Modal share	0.73	0.72	0.72
		Public transport	0.73	0.71	0.71
		Private transport	0.30	0.26	0.29
	Palma	Modal share	146.78	143.59	127.20
		Public transport	57.56	51.93	48.13
		Private transport	4.10	3.11	3.88

As observed in Table 1, the resulting Gini coefficients indicate a high level of inequality in accessibility to parks and cultural equipment. The inequality in accessibility to parks is lower than the inequality in accessibility to cultural equipment due to the high concentration of cultural equipment in the central region of São Paulo when compared to parks distribution. The Gini coefficients are very similar for Saturday, Sunday, and Wednesday, indicating that during the analyzed days the traffic conditions and the transit supply do not have a strong influence on inequalities. The scenario with the lowest inequality was the one in which the entire population uses private transport (hypothetical scenario, since there are cost limitations for using private transport and the traffic database considers that only a part of the population uses private transport), followed by the scenario in which the population uses public transport. The scenario with the highest inequality is the one that considers the modal split of the OD zones in the calculation of the Gini coefficient.

The Palma coefficient reinforces the inequalities found through the Gini coefficient, by indicating that inequalities are in general very high, that the inequality in accessibility to cultural equipment is higher than to parks, and that modal share strongly influences inequalities.

Conclusions

This paper aimed to calculate the accessibility to leisure (parks and cultural equipment) in the municipality of São Paulo and to identify its inequalities considering different class and race population groups. The methodology used in this study seizes the opportunity of the availability of different traffic data sources varying throughout the selected days to generate a multitemporal analysis. Thus, traffic data for public and private transport were used to create transport networks in a GIS environment for 06 (Saturday), 07 (Sunday), and 10 (Wednesday), April 2019. The accessibility inequalities were obtained by stratifying the population according to class and race (high-black, high-white, low-black, and low-white) and calculating the Lorenz curves, Gini, and Palma coefficients for the three analyzed days considering the transport mode.

The results of the analysis indicated a sharp core-periphery division of accessibility to leisure, where the central area concentrates most of the OD zones with high accessibility to leisure, while the peripheral areas are characterized by low accessibility levels. This result is obtained regardless of the day analyzed. The transport mode was also found to strongly influence the resulting accessibility to leisure since accessibility by public transport is substantially lower than the accessibility by private transport.

The accessibility distribution and its inequalities are also analyzed in this study. It revealed that the high-white population is the population group with the highest level of accessibility to leisure. This can be explained by the residential location, the concentration of the leisure opportunities near its residences, and the modal share of this group, which is marked by the predominant use of private transport. The second and the third group in terms of accessibility to leisure is the high-black and the low-white population. The proximity of both groups indicates that even when the black population has the means to live in areas with high accessibility to leisure, its accessibility characteristics are more similar to the low-white population than the high-white population. The population group with the lowest level of accessibility to leisure is the low-black population. This population is, in the Brazilian context, marked by segregation and by residing in areas with poor transport and infrastructure in general.

This study presents some limitations. One limitation is not to weight parks according to their size, service quality, and availability of facilities within it. This limitation should be explored in future work. There are also inherent limitations to the quality of the traffic information used for public and private transport modes. While AVL brings high quality public transport supply information for every hour of the selected days, the private transport traffic data has limited coverage, representing more accurately the main streets of São Paulo. This may lead to an overestimation of accessibility by private transport. Another limitation is the use of stated preferences and individual answers from the 2017 origin and destination survey, which may contain errors, rather than the actual origin and destination of trips.

From a policy perspective, the results show clearly that public transport is much less efficient than private transport to access leisure activities in São Paulo. The situation is of special concern in the peripheral areas, where the public mass rail transport system is poor and there are less leisure opportunities than in the central area. Investments in a more efficient public transport, especially in the peripheral areas of São Paulo, are instrumental to improve accessibility to leisure. Additionally, land-use policies to better distribute leisure opportunities over the municipality could have significant impacts in accessibility to leisure. We hope that the empirical evidence brought by this paper may help shed light on the fact that a major joint policy effort must be made in the pursuit of spatial equity in the access to leisure, on the path to developing more equitable and pleasant cities.

Supplemental Material

Supplemental Material - Unfolding time, race and class inequalities to access leisure

Supplemetary Material for Unfolding time, race and class inequalities to access leisure by Diego B Tomasiello and Mariana Giannotti in Environment and Planning B: Urban Analytics and City Science

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors would like to thank the financial support from the São Paulo Research Foundation (FAPESP) [grant number: 13/07616-7] and the National Council for Scientific and Technological Development (CNPq) [Grant Number: 312774/2020-6].

ORCID iDs

Diego Bogado Tomasiello

Mariana Giannotti

Supplemental Material

Supplemental material for this article is available online

Notes

Diego Bogado Tomasiello is a postdoc researcher in the Center for Engineering, Modeling and Applied Social Sciences at the Federal University of ABC. He holds a bachelor’s degree in Cartographic Engineering from the Federal University of Paraná, a Master’s and Doctoral degree in Transportation Engineering from the Polytechnic Schools of the University of São Paulo with an exchange program at the Department of Infrastructure Engineering from the University of Melbourne. His research focuses on accessibility inequalities, spatial dynamic modeling, and transport networks.

Mariana Giannotti is an assistant Professor at the Polytechnic School of the University of São Paulo, Coordinator of the Laboratory for Geospatial Analysis at the same university and Researcher at the Center for Metropolitan Studies. She holds a Master’s degree in Remote Sensing and Geospatial Analysis from the National Institute for Space Research; and a PhD in Geospatial Analysis from the Polytechnic School of the University of São Paulo, with an exchange program at the Institut National des Sciences Appliquées de Lyon. Her research agenda focuses on socio-spatial inequalities related to urban transport.

References

Batty

(2009) Accessibility: in search of a unified theory. Environment and Planning. B, Planning & Design 36(2): 191–194. DOI: 10.1068/b3602ed

Ben-Akiva

Lerman

(1979) Disaggregate travel and mobility choice models and measures of accessibility. In: Hensher

Sopher

(eds), Behavioural Travel Modelling. Andover, Hants: Croom Helm, 654–679.

Bills

Walker

(2017) Looking beyond the mean for equity analysis: examining distributional impacts of transportation improvements. Transport Policy 54: 61–69. DOI: 10.1016/j.tranpol.2016.08.003

Bittencourt

Giannotti

Marques

(2020) Cumulative (and self-reinforcing) spatial inequalities: interactions between accessibility and segregation in four Brazilian metropolises. Environment and Planning B: Urban Analytics and City Science 48: 1989–2005. DOI: 10.1177/2399808320958426

Caldeira

(2001) City of Walls: Crime, Segregation, and Citizenship in Sao Paulo. Berkeley: University of California Press.

Comber

Brunsdon

Green

(2008) Using a GIS-based network analysis to determine urban greenspace accessibility for different ethnic and religious groups. Landscape and Urban Planning 86(1): 103–114. DOI: 10.1016/j.landurbplan.2008.01.002

Companhia do Metropolitano de São Paulo (2007) Pesquisa Origem e Destino 2007. http://transparencia.metrosp.com.br/dataset/pesquisa-origem-e-destino.

De Vries

Verheij

Groenewegen

, et al. (2003) Natural environments—healthy environments? An exploratory analysis of the relationship between greenspace and health. Environment and Planning A 35(10): 1717–1731. DOI: 10.1068/a35111

Delbosc

Currie

(2011) Using Lorenz curves to assess public transport equity. Journal of Transport Geography 19(6): 1252–1259. DOI: 10.1016/j.jtrangeo.2011.02.008

10.

Foth

Manaugh

El-Geneidy

(2013) Towards equitable transit: examining transit accessibility and social need in Toronto, Canada, 1996–2006. Journal of Transport Geography 29: 1–10. DOI: 10.1016/j.jtrangeo.2012.12.008

11.

Geurs

Van Wee

(2004) Accessibility evaluation of land-use and transport strategies: review and research directions. Journal of Transport Geography 12(2): 127–140. doi: 10.1016/j.jtrangeo.2003.10.005.

12.

Giannotti

Barros

Tomasiello

, et al. (2021) Inequalities in transit accessibility: Contributions from a comparative study between Global South and North metropolitan regions. Cities 109(103016). doi: 10.1016/j.cities.2020.103016.

13.

Gini

(1912) Variability and Mutability, Contribution to the Study of Statistical Distributions and Relations. Studi Economico-Giuricici della R. Universita de Cagliari.

14.

Grahn

Stigsdotter

(2010) The relation between perceived sensory dimensions of urban green space and stress restoration. Landscape and Urban Planning 94(3–4): 264–275. DOI: 10.1016/j.landurbplan.2009.10.012

15.

Gupta

Roy

Luthra

, et al. (2016) GIS based analysis for assessing the accessibility at hierarchical levels of urban green spaces. Urban Forestry & Urban Greening 18: 198–211. DOI: 10.1016/j.jtrangeo.2016.10.012

16.

Hansen

(1959) How accessibility shapes land use. Journal of the American Institute of Planners 25(2): 73–76. DOI: 10.1080/01944365908978307

17.

Hernandez

(2018) Uneven mobilities, uneven opportunities: social distribution of public transport accessibility to jobs and education in Montevideo. Journal of Transport Geography 67: 119–125. DOI: 10.1016/j.jtrangeo.2017.08.017

18.

Hoffimann

Barros

Ribeiro

(2017) Socioeconomic inequalities in green space quality and accessibility—Evidence from a Southern European city. International Journal of Environmental Research and Public Health 14(8): 916. DOI: 10.3390/ijerph14080916

19.

Jin

Kong

Sui

(2019) Uber, public transit, and urban transportation equity: a case study in new york city. The Professional Geographer 71(2): 315–330. DOI: 10.1080/00330124.2018.1531038

20.

Lorenz

Max

O (1905) Methods of measuring the concentration of wealth. Publications of the American statistical association 9(70): 209–219. doi: 10.2307/2276207.

21.

Martens

(2012) Justice in transport as justice in accessibility: applying Walzer’s ’Spheres of Justice’ to the transport sector. Transportation 39(6): 1035–1053. DOI: 10.1007/s11116-012-9388-7

22.

Mohri

Mortazavi

Nassir

(2021) A clustering method for measuring accessibility and equity in public transportation service: case study of Melbourne. Sustainable Cities and Society 74: 103241. DOI: 10.1016/j.scs.2021.103241

23.

Neutens

Schwanen

Witlox

, et al. (2010) Equity of urban service delivery: a comparison of different accessibility measures. Environment and Planning A 42(7): 1613–1635. DOI: 10.1068/a4230

24.

Omer

(2006) Evaluating accessibility using house-level data: a spatial equity perspective. Computers, Environment and Urban Systems 30(3): 254–274. DOI: 10.1016/j.compenvurbsys.2005.06.004

25.

Páez

Scott

Morency

(2012) Measuring accessibility: positive and normative implementations of various accessibility indicators. Journal of Transport Geography 25: 141–153. DOI: 10.1016/j.jtrangeo.2012.03.016

26.

Palma

(2011) Homogeneous middles vs. heterogeneous tails, and the end of the ‘inverted-U’: it’s all about the share of the rich. Development and Change 42(1): 87–153. DOI: 10.1111/j.1467-7660.2011.01694.x

27.

Pereira

(2019) Future accessibility impacts of transport policy scenarios: equity and sensitivity to travel time thresholds for bus rapid transit expansion in Rio de Janeiro. Journal of Transport Geography 74: 321–332.

28.

Pritchard

Tomasiello

Giannotti

, et al. (2019) An international comparison of equity in accessibility to jobs: London, São Paulo and the Randstad. Transport Findings February(7412). doi: 10.32866/7412.

29.

Ricciardi

Xia

Currie

(2015) Exploring public transport equity between separate disadvantaged cohorts: a case study in Perth, Australia. Journal of Transport Geography 43: 111–122.

30.

Rojas

Páez

Barbosa

, et al. (2016) Accessibility to urban green spaces in Chilean cities using adaptive thresholds. Journal of Transport Geography 57: 227–240. DOI: 10.1016/j.jtrangeo.2016.10.012

31.

Siqueira-Gay

Giannotti

Sester

(2019) Learning about spatial inequalities: Capturing the heterogeneity in the urban environment. Journal of Cleaner Production 237: 117732. DOI: 10.1016/j.jclepro.2019.117732

32.

Slovic

Tomasiello

Giannotti

, et al. (2019) The long road to achieving equity: job accessibility restrictions and overlapping inequalities in the city of São Paulo. Journal of Transport Geography 78: 181–193. DOI: 10.1016/j.jtrangeo.2019.06.003

33.

Sullivan

Kuo

Depooter

(2004) The fruit of urban nature: vital neighborhood spaces. Environment and Behavior 36(5): 678–700. DOI: 10.1177/0193841X04264945

34.

Tahmasbi

Mansourianfar

Haghshenas

, et al. (2019) Multimodal accessibility-based equity assessment of urban public facilities distribution. Sustainable Cities and Society 49: 101633. DOI: 10.1016/j.scs.2019.101633

35.

Van Wee

Geurs

(2011) Discussing equity and social exclusion in accessibility evaluations. European Journal of Transport and Infrastructure Research 11(4): 350–367. doi: 10.18757/ejtir.2011.11.4.2940.

36.

Vecchio

Tiznado-Aitken

Hurtubia

(2020) Transport and equity in Latin America: a critical review of socially oriented accessibility assessments. Transport Reviews 40(3): 354–381. DOI: 10.1080/01441647.2020.1711828

37.

Wang

Brown

Liu

(2015) The physical and non-physical factors that influence perceived access to urban parks. Landscape and urban planning 133: 53–66. doi: 10.1016/j.landurbplan.2014.09.007.

38.

Wang

Lan

(2019) Park green spaces, public health and social inequalities: understanding the interrelationships for policy implications. Land Use Policy 83: 66–74. DOI: 10.1016/j.landusepol.2019.01.026

39.

Wei

(2017) Greener urbanization? Changing accessibility to parks in China. Landscape and Urban Planning 157: 542–552. DOI: 10.1016/j.landurbplan.2016.09.004

40.

Xing

Liu

(2018) Measuring spatial disparity in accessibility with a multi-mode method based on park green spaces classification in Wuhan, China. Applied Geography 94: 251–261. DOI: 10.1016/j.apgeog.2018.03.014

41.

Yang

Wang

Zhou

, et al. (2018) Walking accessibility and property prices. Transportation Research Part D: Transport and Environment 62: 551–562. DOI: 10.1016/j.trd.2018.04.001

42.

Yasumoto

Jones

Shimizu

(2014) Longitudinal trends in equity of park accessibility in Yokohama, Japan: an investigation into the role of causal mechanisms. Environment and Planning A 46(3): 682–699. DOI: 10.1068/a45683

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.22 MB