Stockholm Resilience Centre Research for Biosphere Stewardship and Innovation

(1)

Master’s Thesis, 60 ECTS

Social-ecological Resilience for Sustainable Development Master’s programme 2016/18, 120 ECTS

Exploring the role of urban environments for human wellbeing: an analysis of

people’s experiences in Madrid

Paloma Franco Nieto

Stockholm Resilience Centre

Research for Biosphere Stewardship and Innovation

(2)

2

Exploring the role of urban environments for human wellbeing:

an analysis of people’s experiences in Madrid

Paloma Franco Nieto

Master’s thesis – Social-ecological Resilience for Sustainable Development Stockholm Resilience Centre

Stockholm University

Supervisor: Stephan Barthel

Stockholm Resilience Centre Stockholm University

Co-supervisor: Karl Samuelsson University of Gävle

(3)

3

(4)

4

Abstract

If urban planning is to ensure wellbeing while reducing negative environmental impacts, a better understanding on how different urban environments support or hinder wellbeing is needed. This thesis uses softGIS methodology to understand how different urban environments impact people’s experiences in Madrid. An online PPGIS survey collected people’s positive and negative experiences in Madrid (n=400) as well as the perceived environmental qualities (PEQ) for each experience. The thesis applies a psychological approach to the analyses of experiences by applying affordance theory.

For the spatial analysis, the study uses 6 environmental features and analyses the perceived environmental qualities in all of them. The results suggest that social interaction is the main PEQ for having positive experiences and that it is higher in built environments. On the other hand, the presence of nature is reported to be the most important PEQ in positive experiences in nature environments. However, social interaction is the main cause for having negative experience regardless of the environment. The thesis concludes that this method allows to map restorative environments and describe their PEQ providing a useful tool for urban planners to design cities for citizens’ wellbeing. This thesis suggests that in order to achieve sustainability goals in urban areas while ensuring wellbeing, a focus should be put on transforming places with high number of negative experiences by including some nature elements that can reduce the feeling of crowding without eroding the dynamics of these environments. Also, a better distribution of nature environments could improve wellbeing in urban areas. In addition, creating spaces for social interaction should be other priority in urban planning in Madrid, due to the importance it has for people’s wellbeing.

Key words: perceived environmental qualities, restorative environments, wellbeing, experiences, PPGIS, Madrid.

(5)

5

1. Introduction

The aim of the thesis is to explore the linkages between different urban environments and human wellbeing by the analysis of people’s experiences. Wellbeing is an umbrella concept for five dimensions of the state of a person’s life, which are: basic material for a good life, good social relations, health, safety and freedom of choice and action (MA 2005).

With rapid urbanization, urban planning faces the challenge of reducing negative environmental impacts (Grimm et al. 2008; Kennedy et al. 2015) while ensuring the wellbeing of their citizens (Samuelsson et al. 2018). Currently, cities host more than a half of total population. Meanwhile, cities occupying 2% of total land, are responsible for 75% of greenhouse emissions and 78% of total waste (United Nations 2017).

According to the UN (2014), in 2050 cities will host 2.5 billion more of people. With an increasing urban population, designing sustainable cities is in the main Agenda of urban planning (UN Habitat et al. 2016). Nevertheless, because cities are the habitat of the citizens, a work on achieving sustainability also needs to take into account citizens’

wellbeing (UN Habitat 2016).

In order to reduce the negative environmental impact of cities, the compact city model gained a lot of favour as a model of sustainable city (Samuelsson et al. 2018) since it is argued that it is energy efficient and that it occupies less land. The main guiding principle of the compact city model is increasing population density with the goal to reduce travel by car (Newman, 2006) and minimize floor space per capita to heat/cool, and to light up (Jenks et al. 1996; Kennedy et al. 2015). In addition, compaction leaves more space outside the city for biodiversity conservation (Soga et al. 2014).

However, compaction might threaten citizens’ wellbeing due to an increase of sources of stress and mental fatigue (Lederbogen et al. 2011). Living in urban areas is associated with psychosocial diseases such as stress, anxiety or depression (Evans 2003; Bolund and Hunhammar 1999). Such diseases are associated with, among other factors, crowding, dense traffic or pollution (Press et al. 2010). These factors also dishearten urban dwellers from physical activity not providing safe public spaces (WHO, 2016).

Thus, high density environments proposed by the compact city model increase not only sources of stress but also the discourage of physical activity.

(7)

7

On the other hand, there are a large number of studies reporting the benefits of having nature experiences for human wellbeing (Soga et al. 2014; Pretty 2011). Nature experiences (experiences taking place in nature) are good for mental health (e.g.

restoration from stress) and physical health (physical exercise) (Maller et al. 2010;

Wells et al. 2007). Several studies have reported how exposure to nature helps recovering from stress quicker and restoring attention as well as encourages physical activity (Ulrich et al., 1991; Kaplan & Kaplan, 1979; Wells et al. 2007). Thus, nature environments are considered restorative environments (e.g. Hartig et al. 1991).

Restorative environments are “environments that promote wellbeing by supporting their recovery from efforts to meet the demands of everyday life” (Collado et al. 2016). In addition, nature environments have been associated to favour social interaction (Coley et al 1997; Thompson 2002).

Consequently, research emphasizing the importance of nature experiences for human wellbeing challenges the compact city model, since compaction leads to an increase on the demand for restorative environments while it reduces the offer of this type of environments. However, if urban planning is to ensure citizens’ wellbeing, there is a need to understand which environments citizens feel as restorative environments and how these environments support restoration. For that, this thesis empirically analyses people’s experiences in Madrid, Spain. It explores the main perceived environmental qualities (social interaction, nature and built objects) in people’s experiences. The following research questions are formulated:

1) Does the importance of social interaction, nature and built objects vary for people’s experiences in Madrid?

2) Does the importance of social interaction, nature and built objects for people’s experiences vary across the different urban environments in Madrid?

In order to answer these research questions an online PPGIS survey was done to collect people’s experiences in Madrid. Public Participatory Geographical Information System (PPGIS) is a geospatial method for “engaging non-experts to identify the spatial dimensions of social and cultural landscapes” (Brown and Kyttä 2014). This survey is a replication of a survey done in Stockholm which explored the impact of accessibility to different environments on the outcome of people’s experiences. A better understanding on the role of the urban environments on city dwellers wellbeing might allow urban planning to move towards sustainability while ensuring citizens’ wellbeing.

(8)

8

2. Theory and analytical framework

This thesis explores how human-environment interaction leads to wellbeing by the analysis of people’s experiences because experiences are the interface between the environment and wellbeing (Kyttä et al. 2016). Within environmental psychology, place-experience research has put the main focus on individuals, paying little attention to the role of the environment in this relation (Kyttä et al. 2013). However, because human-environment interaction is not a one-way relation but rather a two-ways relation, this thesis applies the perceptual ecological psychology theory of affordances to the place-experience analysis. Affordances are defined as the relation between the abilities of the individual and the perceived properties of the environment (Chemero 2003).

Affordance theory (Gibson, 1979) breaks the dualism in the human-environment interaction and it rather applies an embodied approach in which this interaction is understood as being a match between the environment and the person (Raymond, Giusti, and Barthel 2017). Based on the example Raymond et al (2017) use in their explanation of affordance theory for an embodied approach of cultural ecosystem services, depending on the individual, a steep slope (clue within the environment) might be perceived as a perfect place for mountain-biking, a place for hiking or as a dangerous feature. Hence, affordances are approximately synonymous with the concept of environmental qualities and an experience depends, as it has been exemplified, on both the environmental qualities as directly perceived by the individual, her abilities, and on the culture in which the same individual is embedded (Raymond et al., 2017). Socio- cultural aspects of the environment always have been present in affordance theory (Gibson 1979). Hence affordances are dynamic and coupled human-environment relations (Kaaronen 2017), (where behaviour is probabilistic and actualisation of affordances into various forms of social action may occur when the right social circumstances are present (Clark and Uzzell 2002; Delia and Krasny 2018). This thesis uses environmental qualities in line with affordance theory, hence “the social” together with bio-physical properties form part of environmental qualities, which in combination shape experiences and behaviour (Barker 1968).”

To study the relation between an experience and wellbeing, this thesis applies the psychological approach of stress recovery theory. Stress is the resulting condition from a person-environment interaction that leads to a perceived discrepancy for the individual between what the situation demands and the resources (biological, psychological or

(9)

9

social) the individual has for it (Berto 2014). The assumption of this study is that stress reduction (or the absence of stress) is implicit in a positive experience, thus it assumes that a positive experience is an experience that contribute to reduce stress levels. In this study, wellbeing is conceived as a psychological state of pleasure and relaxation.

Stress recovery theory was first developed by (Ulrich et al 1991) when finding out that people recovered better from stress after seeing pictures/tapes of nature environments rather than pictures/tapes of urban places. They measured stress and stress reduction from self-reported affective states as well as several physiological measures. Several studies have validated this theory with experimental studies (Alvarsson 2001; Hartig et al 2007; Pretty 2011). For instance, Hartig et al (2007) found how the conscious understanding of the psychological restoration benefits of nature triggers ecological behaviour. While nature benefits on recovery from stress have been largely proved by a large amount of studies (see Berto 2014 for an overview on the theory), there is not that much research on how built environments contribute to stress recovery. Nevertheless, several studies have reported that built environments can also offer restorative experiences (Beil and Hanes 2013; Mouratidis 2018, 2017). For instance, Scopelliti and Giuliani (2004) found that both, nature and built environments, have the potential to provide restorative experiences, however the restorative components differ between the two types of environments.

Within urban planning, mapping perceived environmental qualities (affordances) is at the core of research trying to understand what are the aspects of the urban structure that lead to wellbeing. Current research on place-based experiences is applying softGIS methodology to identify what are the qualities in the environment that people perceive.

Kyttä et al (2011) used PPGIS methodology to map environmental qualities of the living environments personally meaningful in Helsinki (Finland). Kyttä et al. (2013) mapped people’s experiences in Helsinki (Finland) to explore meaningful environmental qualities and their accessibility and how they vary in different urban fabric. (Kyttä et al. 2016) studied people’s experiences in Helsinki to analyse the relation between accessibility to different urban environments, the perceived environmental qualities and wellbeing by a questionnaire about different aspects of wellbeing. (Samuelsson et al. 2018) mapped experiences in Stockholm to analyse the impact of accessibility to different environmental features on the outcome of the experience. This thesis builds on the latter study to find empirical evidence about what

(10)

10

is the importance of social interaction, nature and built objects (three groups of environmental qualities) for wellbeing in the different urban environments in Madrid.

3. Study area

Madrid is a dense city which is prospected to continue to grow. Navigating such growth in a sustainable way requires an environmental urban planning that can ensure current and future wellbeing of its citizens. Rapid urbanisation in the last centuries greatly reduced nature in central Madrid - natural environments are in the suburbs of the city.

Within the city there mainly are human designed and fenced nature areas. But those are of great relevance since these are the ones most accessible to citizens and provide citizens with a row of ecosystem services, including important health benefits.

Madrid is the capital of Spain as well as its largest (about 60.000 ha) and most populated (3.2 million inhabitants) city in the country. The city is formed by 21 districts with very different percentage of nature; the ones in the outside parts are greener (more the North than the South located), and the core districts are very compact, with high demographic density and with almost no nature (see Figure 1).

Figure 1. Location of Madrid municipality within Spain. The map of Madrid includes the environmental features used for the analysis of the study. Source of the map of Spain: Natural Earth dataset. Source of the map of Madrid:

CORINE land cover project.

(11)

11

4. Methods

4.1. PPGIS, a method to map place perception

This thesis applies affordance theory to study human-nature interaction. Because affordances are unique for each individual and depend on the perception of the environment, a method that allows to know personal experiences in a microscale level is required to study them (Kyttä et al. 2016). Public Participation Geographical Information Systems (PPGIS) is a “field within geographic information science that focuses on ways the public uses various forms of geospatial technologies to participate in public processes, such as mapping and decision making” (Tulloch 2008). The aim of PPGIS is to enhance social participation processes to generate maps about the cultural landscape to improve the quality of land use decisions (Brown and Kyttä 2014) . Thus, PPGIS represents a good method for the study of affordances since it allows to generate large datasets about people’s experiences (Kyttä et al. 2013).

4.2. PPGIS online survey to collect data of people’s experiences in Madrid In order to answer the research questions, a replication of the online PPGIS survey

‘where is your Stockholm?’ (Giusti, Barthel, Samuelsson, Stockholm University, 2017) was done in Madrid retitled as ‘¿dónde está tu Madrid?’ (‘where is your Madrid?’).

This survey captured people’s experiences in Madrid. Respondents had to pinpoint in a map positive and/or negative experience(s) occurring in their daily routine in Madrid.

After that, they could rate from 1 (not at all) to 10 (very much) how three factors (social interaction, built objects or infrastructure and nature) influenced their experience (Figure 2). There were also questions about the frequency and duration of the experiences. Sociodemographic questions were also included (gender, age, district of residence and nationality). A website (www.dondeestatumadrid.com) was created for people to access the survey. This website showed a description of the project, a link to the survey and the so called ‘mapa de l@s madrileñ@s’ (‘madrileñ@s’ map’) which showed people’s experiences (positive or negative) as points in the map. CARTO programme was used to illustrate people’s experiences on a map which was embedded in the website. The survey was disseminated through social media (Facebook, Twitter and Instagram), personal contacts, pamphlets and local newspapers. Data was collected from October 2017 to February 2018.

(12)

12

Figure 2. Screenshot of the online PPGIS survey after mapping an experience.

4.3. Selection of environmental features

To allow comparison to the study done in Stockholm by Samuelsson et al. (2018), the selected features were based on it (see Table 1¡Error! No se encuentra el origen de la referencia. for the environmental source description). The environmental features used for this study were selected from the CORINE land cover GIS layer (scale 1:100.000).

This GIS layer was developed by the CORINE project (Coordination of Information on the Environment), a project from the European Environment Agency which works on different environmental issues. Natural environments and water bodies were selected as there are a lot of studies highlighting their importance for reducing stress (Pretty 2011).

Major roads were selected as a source of stress, since traffic is associated to stress in urban areas (Evans 2003). Different degrees of urban fabric density were used to analyse how densification might influence people’s experience. Commercial areas were also included in the analysis. Commercial areas differ from urban fabric in the land use they have, i.e. urban fabric density only refers to the density of the built-up areas while commercial areas refer to a specific use certain areas have. There is no overlapping between these features in the CORINE landcover GIS layer. Due to only these features being selected for the analysis, not all the areas within the maps of this study are covered by a land use, which leads to some parts of the map being white. Thus, experiences taking place in areas which land use is not include in this study are excluded from the analysis. For instance, the outside part of Madrid is covered by agricultural and semi-natural areas, which are not analysed in this study.

(13)

13

Table 1. Definition of the environmental features used for the analysis of the experiences. The source of these definitions is CORINE land cover (EEA, 2012). Commercial areas differ from urban fabric density in the land use they have. However, they do not overlap in the GIS layer.

ENVIRONMENTAL FEATURES Nature environments

Green urban areas: areas with vegetation within urban fabric. Includes parks and cemeteries with vegetation.

Forests: vegetation formation composed principally of trees, including shrub and bush understories.

Water bodies

Natural or artificial stretches of water Major roads

Motorways, railways, including associated installations (stations, platforms, embankments). Minimum width of the roads included: 100 m.

Commercial areas

Artificially surfaced areas (with concrete, asphalt, tamacadam, or stablished, e.g. beaten earth) devoid of vegetation, occupy most of the area in question, which also contains buildings and/or vegetated areas.

Urban fabric density (continuous, dense, medium, low)

Continuous: most of the land (> 80 %) is covered by buildings, roads and artificially surfaced area cover almost all the ground. Non-linear areas of vegetation and bare soil are exceptional.

Discontinuous: most of the land is covered by structures. Buildings, roads and artificially surfaced areas associated with vegetated areas and bare soil, which occupy discontinuous but significant surfaces.

- Dense: 50 – 80 % - Medium: 30 – 50 %

4.4. Data transformation and cleaning

As this thesis focuses on experiences happening in different types of environments within the city, to create the nature layer, only green areas which had more than 1 ha of green space were selected, to ensure that the primary characteristic of the space is the nature and the experience of the space is not likely to be affected by other land uses.

Due to potential precision error when respondents mapping experiences, a buffer area around every environmental feature (nature and built environments) was included with QGIS. Different buffer distances were tried before finding a distance considered good enough for avoiding an excess of overlapping of experiences while maintaining a potential mapping precision error. This distance was 50 m around every environmental feature except major roads which had a 20 m buffer. The buffer creates some overlapping of features, leading to some cases in which the same experience belongs to two different environments. In the analysis, these experiences were included in both categories. For instance, if the one experience was included in continuous feature and commercial feature, this experience was in both categories, continuous experience and commercial experience. However, this is not considered a big interference since the main overlapping is between features with similar properties, e.g. discontinuous dense urban fabric and continuous urban fabric. An exception was three small commercial patches within El Retiro, a big central green area. These patches were removed as the main characteristic of the landscape is nature.

(14)

14 4.5. Analysis

The spatial analysis was done with QGIS Desktop 2.18.16. First, in order to classify experiences as happening in a certain environment, e.g. nature experiences, an overlap between the experiences layer (all the collected experiences) and each environmental feature layer was done. The new layers contained the experiences that happen within the patches belonging to a certain environmental feature. Experiences overlapping with several environmental features were treated as belong to each of them. The number of experiences overlapping was calculated pairwise between the environmental features.

The percentage of land covered by each feature was calculated with the area item in QGIS.

To answer research question 1 (does the importance of nature, built objects and social interaction vary in people’s experiences in Madrid?) the data about the three environmental qualities (social interaction, nature and built objects) for the total amount of experiences was used. Their mean and standard deviation was calculated. A one-way ANOVA was performed to compare means of each of these factors. A Tukey’s test was done to see whether factors differed significantly. To answer research question 2 (does the importance of nature, built objects and social interaction vary across the different urban environments in Madrid?) the data about the three environmental qualities (social interaction, nature and built objects) in the different urban environments was used. Their mean and standard deviation was calculated. A one-way ANOVA was performed to compare means of each of these factors across the different environmental features. A Tukey’s test was done to see whether groups differed significantly for each of the factors. Gender was controlled in both cases.

One constraint for the analysis was the discrepancies in the amount of data between the different environments, which did not allow to do hypothesis testing for all the environments. In addition, because the analysis of this study did not look at spatial autocorrelation, the results of this study are indicative rather than conclusive. There could be other factors of the environment that have an impact on the experiences as well as people’s attitudes.

Further research could explore these findings with more sophisticated statistical methods. For example, spatial regression analysis could better account for spatial processes not controlled for in this study to better estimate the effects of social interaction, nature and built object on experiences.

(15)

15

5. Results

5.1. Spatial distribution of the environmental features

The features used for this study occupied 27.04% of the total surface of the municipality of Madrid and mainly occupy its inner part. Continuous urban fabric (land covered by buildings > 80%) occupy 4.11% of Madrid’s total area. Discontinuous dense urban fabric occupies 3.83% and commercial areas occupy 6.86%. As it can be seen in Figure 1, the combination of these three high-density built-up areas occupy almost all the surface of the inner city, while green areas (total area of 8.51%) are mainly in the surrounding neighbourhoods, except for El Retiro, a park that provides some nature to the inner city. Continuous and discontinuous dense urban fabric have similar distribution across the city, predominantly existing in the inner city and the south-west area, without reaching the limits of the municipality. Commercial areas are more scattered, existing in the same areas as continuous and discontinuous dense urban fabric but they also appear in the west and north-west part of the municipality. Discontinuous medium urban fabric (2.01% of the total area) is distributed mainly around the inner neighbourhoods. The main roads (1.67% of the total area) appear as a ring road around the central districts. Water bodies (0.5% of the total surface) are very scarce in the municipality of Madrid, nevertheless there is El Manzanares river, a river that crosses Madrid city. The outside part of the city is mainly white because according to the land cover layer it is agricultural and semi-natural areas, which was not included in analysis.

Also, low and very low urban fabric density were analysed but they were excluded from the analysis due to the lack of experiences in them.

5.2. Responses to the PPGIS survey

There are 390 experiences mapped in the city from 196 respondents. There are almost 3 times more positive experiences (n=308, 78.97%) than negative experiences (n=78, 21.03%). The majority of the collected experiences concentrate in the inner city with a high concentration in the city centre (see Figure 3). Regarding experiences outside the city centre, the experiences in the more outside part of Madrid are mainly positive experiences while most of the negative experiences are scattered over the inner city and around it. Dots of positive experiences are over dots of negative experiences, so some negative experiences might be hidden in the map.

(16)

16

Figure 3. Positive and negative experiences collected with the PPGIS survey in Madrid (the divisions observable in the map are attributed to the different districts of the city). Most of the experiences happen in the inner city with an agglomeration of experiences in the city centre.

63.25% of the respondents were female and 36.14% male. There is an overrepresentation of female in experiences happening in nature environments (66.13%), continuous urban fabric density (71.67%) and in commercial areas (75%).

The other environments analysed have similar representation of both genders. Almost half of the participants had an age between 25 to 34 (43.45%), mapping more experiences than other groups in all the studied environments. The second largest group was the range from 18 to 24 (24.41%) being the second in mapping experiences in all the different environments.

Because it was an online survey, people without access to internet is not represented.

Also, it is possible that mainly active citizens participate in this type of surveys (Rinner and Bird 2009), thus there can be an overrepresentation of a type of population.

However, there is a trade-off between reaching many people and ensuring that the sample is representative of the population. The focus of this thesis was to create and explore a big dataset, therefore an online survey was considered to be useful to reach a

(17)

17

high number of participants. However, it is important to be aware of the type of population that is or is not included when analysing the results from this study.

5.3. Impact of the environment on the experience outcome

Built environments (continuous urban fabric, dense urban fabric, medium urban fabric and commercial areas) gather more experiences (60%) than nature environments (40%).

As it can be seen in Table 2, within built environments, higher densities collected more experiences. Experiences happening in continuous urban fabric (n=170) and commercial areas (n=120) are concentrated in the city centre, while experiences happening in dense areas (n=63) and medium density (n=17) are more scattered within the inner city (see Figure 4). More than half of the positive experiences happening in nature environments (56%) happen in El Retiro. Positive experiences happening in the outside part of the city happen mainly in nature environments. There are few experiences happening close to water bodies (n=11) none of them being negative. Due to the overlapping between the environmental features, there is some overlapping of experiences. The higher overlapping for positive and negative experiences is between continuous urban fabric and commercial areas, followed by continuous urban fabric and dense urban fabric. All experiences close to water bodies overlap with experiences happening in nature environments, furthermore the former experiences are treated as nature experiences.

Table 2. Number and percentage of experiences per environment in relation to the total amount of experiences.

Also, number of experiences overlapping between the different environments analysed Environmental

features Outcome n % Overlapping of experiences per environmental feature Nature Water Commercial Continuous Dense Medium

Nature Positive 135 34.62 - 10 13 10 5 3

Negative 15 58.5 - 0 5 4 2 0

Water Positive 11 2.82 10 - 0 0 0 0

Negative - - 0 - 0 0 0 0

Commercial Positive 81 20.77 13 0 - 48 5 2

Negative 39 10 0 0 - 25 3 1

Continuous Positive 126 32.31 10 0 48 - 30 2

Negative 44 11.28 4 0 25 - 11 0

Dense Positive 44 11.28 5 0 5 30 - 4

Negative 19 4.87 2 0 3 11 - 2

Medium Positive 13 50.7 3 0 2 2 4 -

Negative 4 1.03 0 0 1 0 2 -

(18)

18 +

Figure 4. Positive (green points) and negative (red points) experiences per environmental feature. There are two agglomerations of experiences: nature positive experiences in the only big park in the inner city and positive and negative experiences in the city centre (commercial areas and discontinuous urban fabric). Experiences happening in dense and medium density urban areas are more scattered.

(19)

19

5.4. Impact of social interaction, nature and built objects on the experiences The importance of environmental qualities follows the same order for positive than for negative experiences. The main quality of the environment for experiences to be positive/negative is social interaction (M=7.203 / M=6.69), followed by built objects (M=6.85 / M=5.02) while nature is the less important (M=6.171 / M =3.44). As it can be observed in Table 3, the same happens when analysing the importance of environmental qualities across the different urban environments. Social interaction is the most important factor for having positive/negative experiences in all the built-up environments (M>7.5 for positive experiences and M=7 for negative experiences, except for dense areas M=6.44) and for negative experiences in nature environments (M=6.18). Nature is the highest environmental quality in natural environments and areas close to water bodies (M=7.94 and M=9.14 respectively). For positive experiences, built objects is similar (M=7) in all the environments except for areas close to water where the mean reaches the value of 8.14. For negative experiences the mean is between 4 and 5 for all the environments. Table 3 includes the number of experiences, the mean and standard deviations for all the mentioned experiences. Figure 5, Figure 6 and Figure 7 visualize the amount of data and the mean of the environmental qualities of all the collected experiences, positive experiences across the different urban environments and negative experiences across the different urban environments respectively. Figure 7

(20)

20

Figure 5. Boxplots of the importance of the three studied environmental qualities (built objects, nature and social interaction) in the total collected experiences in Madrid.

Figure 6. Boxplots of the importance of the three studied environmental qualities (built objects, nature and social interaction) in positive experiences in the different urban environments in Madrid.

(21)

21

Figure 7. Boxplots of the importance of the three studied environmental qualities (built objects, nature and social interaction) in negative experiences in the different urban environments in Madrid.

Table 3. Sample size, mean and standard deviation of the three factors from the experience (nature, social interaction, built objects) for all the environments for positive and negative experiences. Social interaction (SI) is the main reason for having positive (M=7.3) and negative (M=6.69) experiences, except in nature and water in which nature is the main factor for positive experiences. Lowest sd for the three factors is in water experiences.

For the rest it varies between 2.5 and 3.5.

Nature Social interaction Built objects

Feature Outcome N M Sd N M Sd N M Sd

Total Positive 193 6.23 3.29 205 7.3 2.52 207 6.95 2.74

Negative 55 3.44 3.16 59 6.69 2.84 58 5.02 3.02

Nature Positive 93 8.03 2.45 93 6.61 2.59 97 6.92 2.82

Negative 9 4 3.57 11 6.18 2.23 10 4.4 2.84

Water Positive 7 9.14 0.69 7 6.43 1.81 8 8.38 1.41

Negative - - - - - - - - -

Commercial Positive 48 3.52 2.50 53 7.85 2.3 53 6.89 2.83

Negative 25 3.48 3.16 28 7.14 2.8 28 4.79 3.1

Continuous Positive 69 3.83 2.91 80 7.81 2.4 77 7.13 2.26

Negative 25 2.92 2.72 29 7.03 2.76 27 4.04 3.1

Dense Positive 26 5.12 3.14 31 8.58 2.01 28 6.71 2.61

Negative 15 3.07 2.99 16 6.44 3.12 15 4.87 2.7

Medium Positive 8 5.63 3.42 9 7.22 2.99 9 6.78 2.68

Negative 3 4.33 3.51 3 7.33 2.08 3 5 2.65

A one-way ANOVA compared the means of social interaction, nature and built objects of all the positive experiences. It was not done for negative experiences, water positive experiences neither medium positive experiences due to the low number of experiences.

This analysis was found to be statistically significant (F(7.104 = 0.000892), df = 2). A post-hoc Tukey’s test showed that social interaction vs nature differed significantly at

(22)

22

0.0001 (p value 0.000356, t value 3.709) and nature vs built objects differed at a significance of 0.5 (p value 0.034573, t value – 2.492).

A one-way ANOVA compared the means of social interaction, nature and built objects across the different environments for positive experiences. It was not done for negative experiences water positive experiences neither medium positive experiences due to the low number of experiences. This analysis was found to be statistically significant for the three of them: social interaction (F(7.104 = 0.000134), df=3), built objects (F(7.104 = 0.000134), df=3) and for nature (F(45.53=<2e-16), df=3).

A post-hoc Tukey’s test showed that taken together, the three categories of urban fabric did not differ from each other, while nature differed from the three of them in all respects. For social interaction, nature vs. commercial groups differed significantly at

<0.05 (p-value 0.016, t value -02.983), nature vs. continuous at <0.01 (p-value 0.0064, t value – 3.167) and nature vs. dense at <0.001 (p-value <0.001, t value – 3.941). For nature, the Tukey’s test showed that nature vs. commercial groups differed significantly at <0.001 (p-value <0.001, t value 9.474), nature vs. continuous at <0.001 (p-value

<0.001, t value 9.880) and nature vs. dense at <0.001 (p-value <0.001, t value 4.907).

For built objects, the Tukey’s test showed that nature vs. commercial groups differed significantly at <0.05 (p-value 0.016, t value – 2.983), nature vs. continuous at <0.01 (p- value <0.00658, t value – 3.267) and nature-dense at <0.001 (p-value <0.001, t value – 3.941).

(23)

23

6. Discussion

Most of the experiences happen in the inner city. More experiences happen in built environments (60%) than in nature environments (40%). The results of this study suggest that social interaction is the main environmental quality for having positive and experiences in Madrid, while nature is the lowest environmental quality for both, positive and negative experiences. According to the results, social interaction is more important in built environments than in nature environments, while experiences happening in nature environments are reported to be more important due to the presence of nature. Built objects are very important in both type of environments.

6.1. The importance of social interaction, nature and built objects in people’s experiences in Madrid

The results of this study align with studies highlighting the importance of social interaction for wellbeing (Carstensen 1995; Nezlek et al. 2002). Social interaction, concretely good social relations, is one of the five dimensions of wellbeing, as defined by MA (2005). In addition, there is a lot of research on how social interaction contributes to mental health (e.g. Martinez Soto et al. 2014; Ono et al. 2011). Having good relations has benefits on depression and helps creating a buffer from stress (Cohen and Wills 1985). Furthermore, it could be argued that environments positively outperforming due to being perceived as good places for social interaction, are environments that promote wellbeing.

The results suggest that the importance of social interaction as a quality of the environments is higher in built environments. The restorative effects of built environments have been increasingly highlighted (e.g. Beil and Hanes 2013; Hartig, Kaiser, and Bowler 2001; Roe 2008). Scopelliti and Giuliani (2004) reported that built environments are also restorative environments, even if the type of restoration offered by built environments is different from the one offered by nature environments. The results of this study indicate that social interaction is better supported by high density environments, due to the considerable higher number of experiences located in these areas in comparison with the other type of urban fabric density areas analysed. This aligns with studies reporting that dense areas are good for social interaction (e.g.

Mardiah 2015). In this study, there is a big agglomeration of positive experiences in the city centre, which might mean that people enjoy places which are lively and full of

(24)

24

people. It can also be because of a wider offer of services such as cafés, restaurants, shops, etc in this area as well as being better connected by public transportation.

On the other hand, the results indicate that social interaction is also the main perceived environmental quality for having negative experiences. Most of the negative experiences are located in dense areas which suggests that crowding could be the negative environmental quality. In addition, negative experiences happening in nature environments are also due to social interaction. The explanatory cause for that could be that due to the lack of nature in Madrid, people go to nature environments seeking calm and relaxation, however the presence of too many people do not allow this type of experience.

Studies focusing on benefits of nature environments for wellbeing frequently report that they are good to promote social interaction (de Bell et al. 2017; Pretty 2011). However, the results of this study suggest that in Madrid the principal motive for having positive experiences in nature environments is the fact that they are rich with nature while social interaction does not stand out in this type of environments. The big difference between the importance of these two environmental qualities in nature environments could be because of the spatial context within nature exists, as Samuelson et al (2018) suggest.

Nature environments in Madrid are mainly in the outside part of the city and only one big green area is located in the inner city. Furthermore, the distribution of nature environments in Madrid does not facilitate their use. This unequal distribution of nature in Madrid could explain the great difference between social interaction and the presence of nature in positive experiences that take place in nature environments. The lack of accessible urban nature in Madrid might trigger that when people go to nature environments they are looking for the presence of nature rather than socializing. In addition, half of the positive experiences happening in nature environments (56%) take place in El Retiro (the big park within the inner city), a fact that highlights the importance of urban nature environments distribution.

According to the results, nature is not perceived as a negative environmental quality in any of the analysed environments. This could be due to nature offering more benefits than harms for wellbeing, however it could also have different reasons. One potential cause is the lack of nature. As commented before, nature areas in Madrid are located in the outside part of the city or they exist as fenced nature within the city, furthermore being in nature can be easily avoided. In addition, dense areas cannot have negative

(25)

25

experiences due to nature, since nature is not present in them even though the lack of nature could be perceived as negative.

In addition, it is interesting the high importance of built objects in nature environments for people’s positive experiences in Madrid. This might mean that people enjoy human- designed nature. It can also mean that people enjoy using facilities offered in this type of environments such as playgrounds for children or areas which provide elements to exercise.

6.2. PPGIS methodology to map restorative environments

This thesis finds softGIS to be a valid method to answer its purpose which is to identify environmental qualities of different urban environments in order to identify the urban restorative environments. While the spatial analysis of experiences contribute identifying the interaction between people and the environment, the PPGIS survey enables to understand what qualities of the environment lead to wellbeing (Kyttä et al.

2016). In addition, applying affordance theory to understand how human-environment interaction leads to wellbeing is understood to be a useful approach as it allows to understand perception of the environment. Also, this thesis finds an experiential approach to be very useful to identify these affordances in relation to a place.

Consequently, the methodology and theoretical framework used in this study is believed to be a very useful tool for urban planning to design cities for wellbeing by taking an experiential approach.

6.3. Implications for urban planning and avenues for future research

This thesis builds on the “affordances as a guiding principle in urban design” suggested by Samuelsson et al. (2018) and identifies the main affordances that could contribute to improve experimental outcome in Madrid. According to this design principle, the findings of this study suggest that one possible way to transform areas with a lot of negative experiences would be the integration of nature elements in these areas, such as big trees. Nature is also highly important for urban wellbeing, however these environments are very unequally distributed within the city. Thus, a better distribution of nature environments might increase the number of experiences in these areas. This is important due to the relevance of nature for wellbeing and the lack of it in urban areas.

These changes, would contribute to promote wellbeing.

This study explores environmental qualities in the different urban environments in the city. However, these environmental qualities are grouped within three categories (social

(26)

26

interaction, nature, built objects). Further research could do a more specific exploration of the environmental qualities and how such environmental qualities shape behaviour.

For instance, it could be explored what is in the environment (cafés, lively, comfortable, etc) that makes it be perceived as good for social interaction. In addition, research on base-placed experiences are mainly in the North of Europe (e.g. Kyttä et al. 2013, 2016;

Samuelsson et al. 2018). These areas have a similar urban structure with high degree of urban natures. These studies highlight the importance of nature for wellbeing in cities.

However, the results of this study suggest that in Madrid, social interaction has more importance than nature for wellbeing. Socio-cultural context and urban structure might be underlying reasons for the difference, furthermore more studies in places with different cultures and urban structures would help to better understand how the environment impact people’s experiences.

(27)

27

7. Conclusion

Urban planning plays an important role in climate change mitigation due to the negative impact cities have on the environment and their projected increase of urban population to 2.5 billion by 2050 (UN, 2014). However, because cities are the habitat of citizens, urban planning needs to take citizens’ wellbeing into account in the work for achieving sustainability (United Nations 2017). In order to achieve that, research on urban planning and research on wellbeing need to be combined, to truly understand what urban environments support or hinder wellbeing and how they do it. This thesis applies softGIS methodology to identify how different urban environments contribute to wellbeing by analysing what are the main perceived environmental qualities for people having positive or negative experiences in them. The results suggest that both nature environments and built environments have the potential to support wellbeing but the causes in doing so differ between them. The study indicates that built environments support wellbeing due to facilitate social interaction while nature environments do it because of being rich with nature. In addition, most of the collected negative experiences happen in built environments and the study indicates that what hinders wellbeing in both type of urban environments is social interaction. This thesis applies

“affordances as guiding principle in urban design”, proposed by Samuelsson et al.

(2018) and suggests that in order to promote wellbeing, urban planning should use an experimental approach to understand what are the environmental qualities that promote or hinder wellbeing. According to the obtained results, this thesis suggests that integrating nature elements in dense built environments could contribute to reduce the number of negative experiences in the city. Also, a better distribution of green areas within the city would help addressing urban stress by creating a buffer to it. In addition, social interaction is the most highlighted environmental quality for positive experiences in Madrid. Furthermore, creating spaces that foster social interaction would contribute to promote wellbeing in Madrid. Applying affordance theory to experiences collected with PPGIS surveys has been proved to be a useful way to study human-environment interaction (Kyttä et al. 2016, 2013). In addition, more studies in places with different urban structures and cultures would contribute to better understand how different urban environments lead to wellbeing.

(28)

28

References

Alvarsson, Jesper. 2001. “Stress Recovery During Exposure To Natural Sounds and Environmental Noise.” Environments, 14.

Barker, R. G. 1968. “Ecological Psychology: Concepts and Methods for Studying the Environment of Human Behavior.,” no. 26: 1980.

Beil, Kurt, and Douglas Hanes. 2013. “The Influence of Urban Natural and Built Environments on Physiological and Psychological Measures of Stress- A Pilot Study.” International Journal of Environmental Research and Public Health 10 (4):

1250–67. https://doi.org/10.3390/ijerph10041250.

Bell, Siân de, Hilary Graham, Stuart Jarvis, and Piran White. 2017. “The Importance of Nature in Mediating Social and Psychological Benefits Associated with Visits to Freshwater Blue Space.” Landscape and Urban Planning 167 (March). Elsevier:

118–27. https://doi.org/10.1016/j.landurbplan.2017.06.003.

Berto, Rita. 2014. “The Role of Nature in Coping with Psycho-Physiological Stress: A Literature Review on Restorativeness.” Behavioral Sciences 4 (4): 394–409.

https://doi.org/10.3390/bs4040394.

Bolund, P, and S Hunhammar. 1999. “Ecosystem Services in Urban Areas.” Ecological Economics 29 (2): 293–301. https://doi.org/10.1016/S0921-8009(99)00013-0.

Brown, Greg, and Marketta Kyttä. 2014. “Key Issues and Research Priorities for Public Participation GIS (PPGIS): A Synthesis Based on Empirical Research.” Applied Geography 46. Elsevier Ltd: 126–36.

https://doi.org/10.1016/j.apgeog.2013.11.004.

Carstensen, Laura L. 1995. “Evidence for a Life-Span Theory of Socioemotional

Selectivity All Use Subject to JSTOR Terms and Conditions Evidence for a Life-Span Theory of Socioemotional Selectivity Partment of Psychology ,.” Current Directions in Psychological Science 4 (5): 151–56. https://doi.org/10.1111/1467-

8721.ep11512261.

Chemero, Anthony. 2003. “An Outline of a Theory of Affordances Anthony.” Ecological Psychology 15 (2): 181–95. https://doi.org/10.1207/S15326969ECO1502_5.

Clark, Charlotte, and David L. Uzzell. 2002. “The Affordances of the Home, Neighbourhood, School and Town Centre for Adolescents.” Journal of Environmental Psychology 22 (1–2): 95–108.

https://doi.org/10.1006/jevp.2001.0242.

Cohen, Sheldon, and Thomas Ashby Wills. 1985. “Stress, Social Support, and the Buffering Hypothesis.” Psychological Bulletin 98 (2): 310–57.

https://doi.org/10.1037/0033-2909.98.2.310.

Coley, Rebekah Levine, Frances E. Kuo, and William C. Sullivan. 1997. “Where Does Community Grow? The Social Context Created by Nature in Urban Public Housing.” Environment and Behavior 29 (4): 468–94.

https://doi.org/10.1177/001391659702900402.

Collado S., Staats H., Corraliza J.A., Hartig T. (2017) Restorative Environments and Health. In: Fleury-Bahi G., Pol E., Navarro O. (eds) Handbook of

Environmental Psychology and Quality of Life Research. International Handbooks of Quality-of-Life. Springer, Cham.

https://doi.org/10.1007/978-3-319-31416-7_7

Delia, Jesse, and Marianne E. Krasny. 2018. “Cultivating Positive Youth Development, Critical Consciousness, and Authentic Care in Urban Environmental Education.”

(29)

29

Frontiers in Psychology 8 (JAN): 1–14. https://doi.org/10.3389/fpsyg.2017.02340.

Duraiappah, Anantha Kumar, Shahid Naeem, Tundi Agardy, Neville J. Ash, H. David Cooper, Sandra Díaz, Daniel P. Faith, et al. 2005. Ecosystems and Human Well- Being. Ecosystems. Vol. 5. https://doi.org/10.1196/annals.1439.003.

Evans, Gary W. 2003. “The Built Environment and Mental Health.” Journal of Urban Health 80 (4): 536–55. https://doi.org/10.1093/jurban/jtg063.

Gibson, J. 1987. The Ecological Approach To Visual Perception. New York:

Psychology Press.

Grimm, Nancy B., Stanley H. Faeth, Nancy E. Golubiewski, Charles L. Redman, Jianguo Wu, Xuemei Bai, and John M. Briggs. 2008. “Global Change and the Ecology of Cities.” Science 319 (5864): 756–60. https://doi.org/10.1126/science.1150195.

Hartig, T., Mang, M., & Evans, G. W. 1991. Restorative effects of natural environment

experiences. Environment and Behavior, 23(1), 3-26.

http://dx.doi.org/10.1177/0013916591231001

Hartig, Terry, Florian G Kaiser, and Peter A Bowler. 2001. “Psychological Restoration in Nature As a Positive Motivation.” Environment and Behavior 33 (4): 590–607.

https://doi.org/10.1177/00139160121973142.

Jenks, Mike, Elizabeth Burton, Katie Williams, and Mike Jenks Elizabeth Burton Katie Williams.

Kaaronen, Roope O, and Environmental Psychology. 2017. “W e i v Re w e i V.”

Kaplan, R. & Kaplan, S. (1989). The Experience of Nature: A psychological perspective. New York: Cambridge University Press.

Kennedy, Christopher A., Iain Stewart, Angelo Facchini, Igor Cersosimo, Renata Mele, Bin Chen, Mariko Uda, et al. 2015. “Energy and Material Flows of Megacities.”

Proceedings of the National Academy of Sciences 112 (19): 5985–90.

https://doi.org/10.1073/pnas.1504315112.

Kyttä, M., Kahila, M. & Broberg, A. Urban Des Int 2011 16: 19.

https://doi.org/10.1057/udi.2010.19

Kyttä, Marketta, Anna Broberg, Mohammed Haybatollahi, and Kaisa Schmidt-Thomé.

2016a. “Urban Happiness: Context-Sensitive Study of the Social Sustainability of Urban Settings.” Environment and Planning B: Planning and Design 43 (1): 34–57.

https://doi.org/10.1177/0265813515600121.

Kyttä, Marketta, Anna Broberg, Tuija Tzoulas, and Kristoffer Snabb. 2013. “Towards Contextually Sensitive Urban Densification: Location-Based SoftGIS Knowledge Revealing Perceived Residential Environmental Quality.” Landscape and Urban Planning 113. Elsevier B.V.: 30–46.

https://doi.org/10.1016/j.landurbplan.2013.01.008.

Lederbogen, Florian, Peter Kirsch, Leila Haddad, Fabian Streit, Heike Tost, Philipp Schuch, Stefan Wüst, et al. 2011. “City Living and Urban Upbringing Affect Neural Social Stress Processing in Humans.” Nature 474 (7352). Nature Publishing Group:

498–501. https://doi.org/10.1038/nature10190.

Maller, Cecily, Mardie Townsend, Lawrence St Leger, Claire Henderson-wilson, Anita Pryor, Lauren Prosser, and Megan Moore. 2010. “Healthy Parks , Healthy People : The Health Benefits of Contact with Nature in a Park Context.” The George Wright Forum 26 (2): 51–83. https://doi.org/10.2307/43598108.

Mardiah, Wan Mohd Rani Wan Nurul. 2015. “Compact Urban Form for Sociability in Urban Neighbourhoods.” International Journal of Social Science and Humanity 5

(30)

30

(10): 822–26. https://doi.org/10.7763/IJSSH.2015.V5.564.

Martinez Soto, J., M. Nanni, L. Gonzales-Santos, E. Pasaye, and F. Barrios. 2014.

“Neural Responses to Restorative Environments: An Eye Tracking and FMRI Study.” 2014 ANFA Conference, no. September: 62–63.

https://doi.org/10.13140/2.1.1340.5446.

Matteo Giusti, Stephan Barthel, Karl Samuelsson. Stockholms universitet, Stockholm Resilience Centre (2017). Var är ditt Stockholm-

bakgrundsinformation. Svensk Nationell Datatjänst. Version 1.0.

https://doi.org/10.5878/002916

Mouratidis, Kostas. 2017. “Built Environment and Social Well-Being: How Does Urban Form Affect Social Life and Personal Relationships?” Cities 74 (March). Elsevier: 7–

20. https://doi.org/10.1016/j.cities.2017.10.020.

———. 2018. “Rethinking How Built Environments Influence Subjective Well-Being: A New Conceptual Framework.” Journal of Urbanism 11 (1). Routledge: 24–40.

https://doi.org/10.1080/17549175.2017.1310749.

Newman, P. (2006). The environmental impact of cities. Environment and Urbanization, 18(2), 275–295.

http://dx.doi.org/10.1177/0956247806069599.

Nezlek, John B., Deborah S. Richardson, Laura R. Green, and Elizabeth C. Schatten- Jones. 2002. “Psychological Well-Being and Day-to-Day Social Interaction among Older Adults.” Personal Relationships 9 (1): 57–71. https://doi.org/10.1111/1475- 6811.00004.

Ono, Eisuke, Takayuki Nozawa, Taiki Ogata, Masanari Motohashi, Naoki Higo, Tetsuro Kobayashi, Kunihiro Ishikawa, Koji Ara, Kazuo Yano, and Yoshihiro Miyake. 2011.

“Relationship between Social Interaction and Mental Health.” 2011 IEEE/SICE International Symposium on System Integration, SII 2011, no. November 2015:

246–49. https://doi.org/10.1109/SII.2011.6147454.

Press, W H O, Avenue Appia, W H O Press, World Health Organi-, and WHO. 2010.

“Unmasking and Overcoming Health Inequities in Urban Settings.” Evolutionary Ecology 10: 1–145. https://doi.org/10.1373/clinchem.2011.163634.

Pretty, J. 2011. “Chapter 23: Health Values from Ecosystems.” UK National Ecosystem Assessment: Technical Report, 1–48.

https://doi.org/papers3://publication/uuid/0CF9310E-4AF6-44F1-8F00- A49C3FACD128.

Raymond, Christopher M., Matteo Giusti, and Stephan Barthel. 2017. “An Embodied Perspective on the Co-Production of Cultural Ecosystem Services: Toward Embodied Ecosystems.” Journal of Environmental Planning and Management 0568 (June). Taylor & Francis: 1–22.

https://doi.org/10.1080/09640568.2017.1312300.

Rinner, Claus, and Michelle Bird. 2009. “Evaluating Community Engagement through Argumentation Maps - A Public Participation GIS Case Study.” Environment and Planning B: Planning and Design 36 (4): 588–601.

https://doi.org/10.1068/b34084.

Roe, Jenny. 2008. “The Restorative Power of Natural and Built Environments,” no.

September: 339.

http://www.ros.hw.ac.uk/bitstream/handle/10399/2250/RoeJ_0908_sbe.pdf?seq uence=1&isAllowed=y.

(31)

31

ROGER S. ULRICH, ROBERT F. SIMONSt, BARBARA D. LOSITOt, EVELYN FIORITOt, MARK A. MILESt and MICHAEL ZELSONt. 1991. “STRESS R E C O V E R Y D U R I N G E X P O S U R E TO N A T U R A L AND URBAN ENVIRONMENTS 1,” 201–30.

Samuelsson, Karl, Matteo Giusti, Garry D. Peterson, Ann Legeby, S. Anders Brandt, and Stephan Barthel. 2018. “Impact of Environment on People’s Everyday Experiences in Stockholm.” Landscape and Urban Planning 171 (July 2017). Elsevier: 7–17.

https://doi.org/10.1016/j.landurbplan.2017.11.009.

Scopelliti, Massimiliano, and M. Vittoria Giuliani. 2004. “Choosing Restorative Environments across the Lifespan: A Matter of Place Experience.” Journal of Environmental Psychology 24 (4): 423–37.

https://doi.org/10.1016/j.jenvp.2004.11.002.

Soga, Masashi, Yuichi Yamaura, Shinsuke Koike, and Kevin J. Gaston. 2014. “Land Sharing vs. Land Sparing: Does the Compact City Reconcile Urban Development and Biodiversity Conservation?” Journal of Applied Ecology 51 (5): 1378–86.

https://doi.org/10.1111/1365-2664.12280.

Thompson, Catharine Ward. 2002. “Urban Open Space in the 21st Century.” Landscape and Urban Planning 60 (2): 59–72. https://doi.org/10.1016/S0169-

2046(02)00059-2.

Tulloch, David L. 2008. “Is VGI Participation? From Vernal Pools to Video Games.”

GeoJournal 72 (3–4): 161–71. https://doi.org/10.1007/s10708-008-9185-1.

UN Habitat, WHO, UNISDR, UN Women, UNEP, UNDP, and UNESCO. 2016. “SDG Goal 11 - Monitoring Framework. A Guide to Assist National and Local Governments to Monitor and Report on SDG Goal 11 Indicators.,” 56. https://unhabitat.org/wp- content/uploads/2016/02/SDG-Goal 11 Monitoring Framework 25-02-16.pdf.

United Nations. 2014. “World Urbanization Prospects.” Undesa.

https://doi.org/10.4054/DemRes.2005.12.9.

WHO. 2016. “The World’s Cities in 2016: Data Booklet.” Economic and Social Affair, 29.

https://doi.org/10.18356/8519891f-en.

WHO. 2017. New Urban Agenda. Conference on Housing and Sustainable Urban Development (Habitat III). https://doi.org/ISBN: 978-92-1-132757-1.

Wells, Nancy M, Susan P Ashdown, Elizabeth H S Davies, and F D Cowett. 2007.

“Environment, Design, and Obesity,” 6–33.