Modelling and Measuring Quality of Urban Life : Housing, Neighbourhood, Transport and Job

Modelling and measuring quality of urban life : housing, neighbourhood, transport and job

Citation for published version (APA): Aminian, L. (2019). Modelling and measuring quality of urban life : housing, neighbourhood, transport and job. Technische Universiteit Eindhoven.

Document status and date: Published: 30/04/2019

Document Version: Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website. • The final author version and the galley proof are versions of the publication after peer review. • The final published version features the final layout of the paper including the volume, issue and page numbers. Link to publication

General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement: www.tue.nl/taverne

Take down policy If you believe that this document breaches copyright please contact us at: [email protected] providing details and we will investigate your claim.

Download date: 07. Oct. 2021

Modelling and Measuring Quality of Urban Life

Housing, Neighborhood, Transport and Job

PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Technische Universiteit Eindhoven, op gezag van de rector magnificus prof.dr.ir. F.P.T. Baaijens, voor een commissie aangewezen door het College voor Promoties, in het openbaar te verdedigen op dinsdag 30 april 2019 om 11:00 uur

door

Lida Aminian

geboren te Teheran, Iran

Dit proefschrift is goedgekeurd door de promotoren en de samenstelling van de promotiecommissie is als volgt:

voorzitter: prof.dr.ir. E.S.M.Nelissen

1e promotor: prof.dr. H.J.P. Timmermans

2e promotor: prof.dr. S. Rasouli leden: prof.dr. D.F. Ettema (Universiteit Utrecht)

prof.dr.ir. B.de Vries

dr.ir. A.D.A.M. Kemperman

Het onderzoek of ontwerp dat in dit proefschrift wordt beschreven is uitgevoerd in overeenstemming met de TU/e Gedragscode Wetenschapsbeoefening.

Modelling and Measuring Quality of Urban Life

Housing, Neighbourhood, Transport and Job

iii

A catalogue record is available from the Eindhoven University of Technology Library

ISBN: 978-90-386-4739-5 NUR: 955

Cover design: Lida Aminian, Reza Saberi photograph: Lida Aminian Printed by the Eindhoven University Press, Eindhoven, The Netherlands Published as issue 268 in de Bouwstenen series of the faculty of Architecture, Building and Planning of the Eindhoven University of Technology

Copyright© Lida Aminian, 2019 All rights reserved. No part of this document may be photocopied, reproduced, stored, in a retrieval system, or transmitted, in any from or by any means whether, electronic, mechanical, or otherwise without the prior written permission of the author.

“Yesterday I was clever, so I wanted to change the world. Today, I am wise so I am changing myself.” Rumi

Preface

Now since this journey is coming to end, I can see it more clearly with all its ups and downs that made me physically and emotionally more resilient compared to the past. Taking a step back in time, I clearly remember the day I received the prompt reply from Prof. Harry Timmermans in response to my email and my interest in doing research in his department. The main question he asked me after looking at my CV was: ” You are not trained as a researcher, so what are the challenges?”. I knew, doing PhD in urban systems and planning would not be an easy route to take for someone with an architecture background, but at that time I was not totally aware of all the dimensions of his question. Now, at the end of this scenario, I can reply to his question differently as I touched all those challenges and even more.

The main challenge is adopting to a new lifestyle. As an architect, I used to work in a team, have a lot of field work, social interactions and discussions with my team members. However, doing research needs a lot of focus, so the isolation is inevitable. Although some scholars love the feeling of autonomy this gives, some prefer more contacts. This is specially more challenging for those whose social circle is relatively limited due to long distances to family and close friends. Learning the strategies to adopt to this new life-style has probably been the biggest milestone in my personal growth during this episode of my life.

The second challenge for someone with a background in architecture and design is to learn the opposite way of presenting ideas. While architecture students learn to develop their ideas and ambiguity can be part of an innovative design, in academic research you should write your ideas in an consistent and clear way, as accurate as possible. The third challenge is related to the nature of the urban planning field. People from almost all engineering fields, geography, social science, economics and applied sciences can contribute to urban planning. However, the most confusing part is that in the Netherlands (and probably some other countries) all of them gathered under the flag of architecture in the faculty of built environment, while their focus is quite different. This can easily deviate one’s initial plan and put one in a totally new world.

vii

Despite all these challenges in the beginning, at some point I realized that I can benefit a lot from my background and the new expertise by formulating new types of interdisciplinary research, bridging design and behavioral science. Working on the “Quality of Urban Life” topic gave me the opportunity to improve this potential. The topic was really interesting in a sense that it connected me to two different directions, one was happiness and well-being and the other concerned the various domains of the built environment. For the first one, I had to conduct an extensive literature study, which was very informative for me and the insights were useful for my own life. It is very rare that you could use the knowledge of your PhD research directly for your own life, though what I learnt is beyond what is reflected in this book. For the second direction, I had to learn quantitative modelling, survey design, data collection etc., which enable me to understand the foundation of research in many different fields. Now, I am happy to spend these years extending and developing my experience in this field, a field with a bright future and so many opportunities to grow.

Although doing a PhD research is an individual activity, the inspiration, motivation and help of others are key success factors. Hereby, I would like to express my gratitude and appreciation to people who have supported me directly and indirectly during this critical period.

Firstly, my greatest thank goes to Prof. Harry Timmermans for his inspirational guidance and patience. His optimism and cooperative attitude have attracted many students from different corners of world and made a very international department. Thank you Harry for all these years, I learnt a lot from you, the way you commit to your work, always having a plan for the future, and your leadership, all were incredibly instructive for me.

I would also like to thank Prof. Soora Rasouli who joined the project in the last stage, but her insightful comments on my thesis enriched my experience and promoted the quality of my dissertation from different angles.

My appreciation also goes to the members of my dissertation committee, prof. de Vries, Dr. Kemperman and prof. Ettema for their valuable time in reviewing the manuscript. Thank you all for your motivational comments. I really enjoyed reading how you described my project. I also like to thank other members of our department: Aloys, Peter, Mandy, Marielle and Joran for their valuable support during this period.

One of the challenging phases of my PhD was data collection, and without help from others it was almost impossible to collect data in the Netherlands. I am

viii

appreciative to Astrid Kemperman, Margot Jaspars, Dujuan Yang and Mark van Glabbeek. I also like to thank the respondents of my survey whose patience and effort in completing such a long survey provided me a very high quality data set, and I hope they may profit from my findings.

Knowing that social aspects of life play a critical role in well-being, I am grateful to all those who made this experience less isolated and more joyful, my friends inside and outside the university for all the gatherings, coffee time, sport activities and chats. I would like to thank my vibrant colleagues Bilin, Bobin, Calvin, Dujuan, Elaine, Eleni, Feixiong, Guangde, Iphi, Marianna, Marielle, Pauline, Rainbow, Robert, Seheon, Sunghoon, Valeria, Wen, Widi, Xiaochen, Yanan, and Zahra (Sareh). Special thanks go to Elaine, Wen and Widi for sharing their critical live moments and experiences with me during this journey.

At a personal level, doing a PhD is a big sacrifice, particularly when it implies long distance from the loved ones. At some point, I have come to believe that emotional resilience is the greatest ability I gained during this period. I had to disappoint my parents several times when I postponed my trips to visit them to finish this dissertation. Nevertheless, without their unconditional support, love and care it was impossible to reach the end of this tunnel. I owe my deepest and heartfelt gratitude to you my dearest, and this dissertation is dedicated to you. I should thank my best friend, a person who grew up with me and all my childhood memories are full of her presence, my sister Leila. Thanks to you and Hosein for the emotional support, listening to my worries, sending us gifts and flowers and visiting us in NL. Thanks also go to my older sister Farimah for her advices and being an inspiration for me. I would also like to extend my thanks to my family in law for their frequent visits, which made our summers more exciting and fun.

Last but not least, my special appreciation goes to my life partner Reza, whose encouragement, support and assistance from the beginning to conclusion enabled me to accomplish this project. We have not spent a day without sharing our thoughts, good and bad moments. My dear companion: I feel very lucky to have you in my life.

Lida Aminian

December 2018

Contents

Preface ...... vii Contents ...... xi List of Figures ...... xv List of Tables ...... xvii 1 Introduction ...... 1 1.1 Background and motivation ...... 1 1.2 Research objectives ...... 3 1.3 Thesis outlines ...... 4 2 Concept and literature review ...... 7 2.1 Introduction ...... 7 2.2 QOL (well–being) ...... 8 2.2.1 Background ...... 8 2.2.2 Theory and measurement ...... 9 2.2.3 Bottom-up spillover theory ...... 11 2.2.4 QOL in environmental psychology: the concept of “person-place- activity” ...... 11 2.3 Quality of Urban Life (QOUL) ...... 13 2.4 Conceptual framework ...... 14 2.5 Conclusions ...... 19 3 Data collection and descriptive statistics ...... 21 3.1 Introduction ...... 21 xi

3.2 Survey design ...... 22 3.2.1 Data consideration ...... 22 3.2.2 Survey instrument ...... 22 3.2.3 The questionnaire ...... 22 3.3 Study areas...... 28 3.4 Sample selection and characteristics ...... 29 3.5 Conclusions ...... 38 4 Housing well-being ...... 39 4.1 Introduction ...... 39 4.2 Literature review ...... 40 4.2.1 A place called house (terminology and definitions) ...... 40 4.2.2 Theoretical concepts ...... 42 4.2.3 Factors determining housing satisfaction (Indicator sets in housing research) ...... 43 4.3 Multiple regression analysis ...... 48 4.3.1 Interpretation ...... 50 4.4 Analysis and results: Path analysis ...... 52 4.4.1 Conceptualization ...... 52 4.4.2 Path analysis for housing satisfaction ...... 53 4.4.3 Interpretation and discussion ...... 57 4.5 Conclusions and discussion ...... 59 5 Neighborhood well-being ...... 61 5.1 Introduction ...... 61 5.2 Literature review ...... 63 5.2.1 A place called neighborhood (terminology and definitions) ...... 63 5.2.2 Research background ...... 64 5.2.3 Theoretical concepts ...... 65 5.3 Factors determining neighbourhood satisfaction ...... 66 5.3.1 Individuals’ socio-demographic characteristics...... 66 5.3.2 Social environment characteristics ...... 67 5.3.3 Physical and environmental characteristics ...... 69 5.3.4 Physical activity satisfaction ...... 69 5.3.5 Satisfaction with services in neighborhood ...... 70 5.4 Analysis and results: Multiple regression analysis ...... 70 5.4.1 Interpretation ...... 71 5.5 Analysis and results: Structural equation modelling (SEM) ...... 76 5.5.1 Conceptualization ...... 77 5.5.2 SEM model for neighborhood satisfaction ...... 79 5.5.3 Interpretation and discussion ...... 87 5.6 Conclusions and discussion ...... 89

xii

6 Transport well-being...... 91 6.1 Introduction ...... 91 6.2 Research background ...... 92 6.3 Factors determining travel satisfaction ...... 93 6.3.1 Individuals’ socio-demographic and household characteristics ... 93 6.3.2 The built environment ...... 94 6.3.3 Travel mode choice ...... 95 6.3.4 Time, the role of rush hours ...... 96 6.3.5 Trip attributes satisfaction ...... 96 6.4 Analysis and results: Multiple regression analysis ...... 99 6.4.1 Interpretation ...... 100 6.5 Analysis and results: Structural equation modelling (SEM) ...... 104 6.5.1 Conceptualization ...... 104 6.5.2 SEM model for transport satisfaction ...... 105 6.6 Interpretation ...... 110 6.6.1 Effects of travel characteristics ...... 110 6.6.2 Effects of socio-demographic variables ...... 112 6.6.3 Effects of built environment ...... 113 6.7 Conclusions and discussion...... 114 7 Work well-being ...... 115 7.1 Introduction ...... 115 7.2 Literature review ...... 116 7.2.1 Activity called job (terminology and definitions) ...... 116 7.2.2 Research background ...... 117 7.2.3 Theoretical concepts ...... 118 7.3 Factors determining work satisfaction...... 119 7.3.1 Individuals’ socio-demographic factors ...... 119 7.3.2 Employment status (Part time vs full time) ...... 121 7.3.3 Work conditions ...... 122 7.3.4 Workspace ...... 124 7.3.5 Work location ...... 126 7.4 Analysis and results: Multiple regression analysis ...... 128 7.4.1 Interpretation ...... 128 7.5 Analysis and results: Path analysis ...... 131 7.5.1 Conceptualization ...... 132 7.5.2 Path analysis model for work satisfaction ...... 132 7.5.3 Interpretation and discussion ...... 135 7.6 Conclusion and discussion ...... 142 8 Quality Of Urban Life (QOUL) ...... 143 8.1 Introduction ...... 143

xiii

8.2 Literature review ...... 144 8.2.1 Domain-based QOL ...... 144 8.2.2 Factors determining QOL ...... 145 8.3 Analysis and results: Multiple regression analysis ...... 149 8.3.1 Interpretation ...... 152 8.4 Analysis and results: Path analysis ...... 152 8.4.1 Conceptualization ...... 152 8.4.2 Path analysis model for QOUL ...... 153 8.5 Interpretation...... 158 8.5.1 The role of socio-demographic attributes on QOUL ...... 158 8.5.2 The influence of life domains on QOUL ...... 159 8.6 Conclusions and discussion ...... 162 9 Conclusions and Future work ...... 165 9.1 Summary ...... 165 9.2 Contribution and Managerial implications ...... 170 9.3 Discussion and direction of future research ...... 172 References ...... 175 Appendix: QOUL Survey ...... 199 Authors Index ...... 215 Subject Index ...... 221 Summary ...... 225 Curriculum Vitae ...... 229

xiv

List of Figures

Figure 2.1 Life domains in relation to person-place- activity concept ...... 15 Figure 2.2 The conceptual framework ...... 20 Figure 3.1 Flyer of the survey ...... 23 Figure 3.2 An example of the final page of neighborhood module ...... 27 Figure 3.3 Location of the four neighborhoods in Eindhoven 1: StrijpS, 2: Schoot, 3: Villapark, 4: Doornakkers ...... 29 Figure 3.4 Local dwelling streets of 1: StrijpS, 2: Schoot, 3: Villapark, 4: Doornakkers 30 Figure 3.5 Chart of household income in study areas ...... 33 Figure 4.1 The Histogram of the overall housing satisfaction ...... 49 Figure 4.2 Normal Probability Plot (P-P) of the regression standardized residual dependent variable for overall house satisfaction ...... 49 Figure 4.3 The main categories of determinants of housing satisfaction ...... 53 Figure 4.4 The conceptual model ...... 54 Figure 4.5 Results of the path analysis for house satisfaction ...... 55

Figure 5.1 The Histogram of the overall neighborhood satisfaction ...... 72 Figure 5.2 Normal Probability Plot (P-P) of the regression standardized ...... 72 Figure 5.3 The conceptual model ...... 78 Figure 5.4 Results of SEM of the modified model with standardized coefficients for neighborhood satisfaction ...... 86 Figure 6.1 The Histogram of the overall travel satisfaction ...... 101 Figure 6.2 Normal Probability Plot (P-P) of the regression standardized residual dependent variable for overall travel satisfaction ...... 101 Figure 6.3 The conceptual model ...... 106 Figure 6.4 Results of the SEM model with standardized coefficients for transport satisfaction ...... 111 Figure 7.1 The Histogram of the overall work satisfaction ...... 129 Figure 7.2 Normal Probability Plot (P-P) of the regression standardized residual dependent variable for overall work satisfaction...... 129 Figure 7.3 The main determinants of the work satisfaction used in this study ...... 131 Figure 7.4 The conceptual model ...... 133 Figure 7.5 Results of the path analysis for work satisfaction ...... 137 Figure 8.1 The Histogram of the overall QOUL ...... 150 Figure 8.2 Normal Probability Plot (P-P) of the regression standardized residual dependent variable for Overall QOUL ...... 150 Figure 8.3 The conceptual model ...... 153 Figure 8.4 Results of path analysis for QOUL ...... 160

xvi

List of Tables

Table 2.1 Integrative theories of QOL ...... 10 Table 2.2 Life domains in QOL studies ...... 12 Table 2.3 Summary of recent QOUL studies ...... 16 Table 3.1 Number of respondents per area ...... 23 Table 3.2 General situation of the four neighborhoods ...... 31 Table 3.3 Description of social-demographic data (N=209) ...... 32 Table 3.4 Descriptive statistics of life domains satisfaction ...... 34 Table 3.5 Descriptive statistics of satisfaction with housing characteristics ...... 34 Table 3.6 Description of housing data, and categories of categorical variables ...... 35 Table 3.7 Descriptive statistics of satisfaction with neighborhood characteristics ...... 37 Table 3.8 Descriptive statistics of satisfaction with transport characteristics ...... 37 Table 3.9 Descriptive statistics of satisfaction with work characteristics ...... 38

xvii

Table 4.1 Different literature regarding the major contributors to the house satisfaction ...... 47 Table 4.2 Goodness-of-fit of the regression model of overall house satisfaction ...... 51 Table 4.3 The estimated parameters of the house satisfaction using multiple regression analysis ...... 51 Table 4.4 Goodness of the fit of the model ...... 56 Table 4.5 Standardized path estimates of direct, indirect and total effects ...... 58 Table 5.1 Goodness-of-fit of the regression model of overall neighborhood satisfaction ...... 73 Table 5.2 The estimated parameters of the neighborhood satisfaction using multiple regression analysis ...... 73 Table 5.3 Goodness of the fit of the initial model ...... 81 Table 5.4 Goodness of the fit of the improved model...... 81 Table 5.5 Standardized path estimates of direct, indirect and total effects ...... 83 Table 5.6 Relationships among the latent and observed variables (direct effects ...... 84 Table 5.7 Relationships among the latent variables ...... 85 Table 6.1 Goodness-of-fit of the regression model of overall travel satisfaction ...... 102 Table 6.2 The estimated parameters of the transport satisfaction using multiple regression analysis ...... 102 Table 6.3 Goodness of the fit of the model ...... 107 Table 6.4 Standardized path estimates of direct, indirect and total effects on overall travel satisfaction...... 108 Table 6.5 Relationships among the latent and observed variables ...... 109 Table 7.1 Major attributes in the selected recent literature ...... 120 Table 7.2 Goodness-of-fit of the regression model for overall work satisfaction ...... 130 Table 7.3 The estimated parameters of the work satisfaction using multiple regression analysis ...... 130 Table 7.4 Goodness of the fit for the path analysis model ...... 134 Table 7.5 Standardized path estimates of direct, indirect and total effects ...... 138 Table 8.1 Goodness-of-fit of the regression model for overall QOUL ...... 151

xviii

Table 8.2 The estimated parameters of the overall QOUL using multiple regression analysis ...... 151 Table 8.3 Goodness of the fit of the model ...... 155 Table 8.4 Standardized path estimates of direct, indirect and total effects ...... 156

xix

Introduction

1.1 Background and motivation

Throughout history, philosophers attempted to define notions of good life or happy life, as every act of human being is directed toward seeking happiness and enhancing Quality Of Life (QOL). In its philosophical definition QOL or sense of well-being is a state of mind that is related to fulfillments of human capacities (Eudaimonia) and pleasure attainment (Hedonia). Beyond the philosophical foundations of the concept, in its core QOL is a universal notion that motivates human behavior and all choices in the life stages. Therefore as Aristotle articulated, it is about doing well and living well. With civilization flourishing, the concept evolved and outlined many new perspectives that reflect positive and negative features of life. These features are now the focal points of wide range of disciplines from psychology to sociology, healthcare, public policy, environmental studies, urban studies etc.

Introduction

Being home to majority of world’s population, cities are key players in providing good life for citizens. So, it is not surprising that more than ever, the QOL debate shifted toward living situations in urban settings, and exploring whether human in particular places such as a city feel happier or not. For many years, social scientists have been examining the role of urban issues such as crime rate or poverty on the sense of well-being, and thereby the role of physical aspects have been usually underestimated or ignored. From 1970 onwards, a major attention has been paid to the role of place characteristics in generating satisfaction. And, basically a new stream of studies has been focused on the role of city design on preventing crime and other issues in major cities. Investigating the capacities of a place in making people happy and leading a desirable life, then extended to abilities of cities in hosting activities that potentially offered to citizens. Following that, cities were ranked based on the objective facilities and standard of living. However, questions such as “Do cities with highest objective characteristics host happiest people?” reflected the need to explore subjective assessments of citizens. To develop understanding regarding city life and subjective well-being of residents, urban scientists borrowed theories and approaches such as domain-specific (life domains) from other fields. Thus, physical domains included to other key domains that were typically measured in subjective well-being studies, and different aspects of cities evaluated and measured quantitatively via surveys.

In addition to that, as judgments of people are highly influenced by individuals’ needs and constraints, socio-demographic information of individuals were also included into subjective well-being surveys. However, despite the importance of quality of urban life (QOUL) studies, the empirical studies that have been explicitly focused on micro level subjective QOL in cities are very limited. Also, among the available studies certain aspects are lacked or ignored. First, QOUL surveys are usually very similar and ask respondents some typical questions, while the concept is very context dependent and to capture specific characteristics of place, particular attention to context is needed. Therefore, indicators in one study may not be relevant to other studies. Second, asking one question regarding a particular domain may not fully capture the respondents view regarding that domain. Also, it may not provide information about the drivers of satisfaction or dissatisfaction, which are important information for policy makers. For instance, the typical question “how satisfied are you regarding your transportation?” cannot reflect what kind of trip and which particular aspects of trips can trigger the answer to this question. In addition, it may be challenging for respondents to judge one

Introduction life domain based on only one question, so some pre-questions are needed to guide the respondents. To prevent such issues it makes sense to ask respondents detail questions regarding their activities in their daily life, and also measure suitability of places in generating satisfaction, using sets of indicators. However, as the concept is very broad, detailed questions in each domain may make the survey very long, which in turn cause data collection to become challenging. That’s why most of QOUL studies undertake by governmental organizations and the rest focus on one or couple of domains such as housing well-being or travel well-being.

This research has been motivated by the above-mentioned gaps in QOUL studies, and aims to study the key issues affecting micro-level subjective well-being. Moreover, the research tends to investigate subjective well-being with explicit focus on the role of the physical and urban environment, divided into four main domains, namely housing well-being, neighborhood well-being, transport well-being and job well-being. This helps first to determine the main aspects influencing each specific domain, and then the extent to which overall satisfaction with that domain may influence overall QOUL. The contribution of this study is to propose an integral theoretical framework for the analysis of QOUL, using spillover theory and domain-specific approach. All the relationships among various indicators, sub-domains and domains are examined using different statistical methods.

1.2 Research objectives

In light of the aforementioned motivations, the main objectives of this study are formulated as:

1. Develop our understanding of subjective well-being1 and QOUL.

2. Examine the extent to which satisfaction with urban characteristics influence the QOUL subjectively experienced by individuals.

3. Investigate the corresponding effects of selected urban life domains (housing, neighborhood, transport and job well-being), and other life domains (i.e. social, leisure, health, financial and family well-being) on each other and QOUL.

4. Examine the strength of the link between objective and subjective indicators of selected urban life domains.

1 Subjective well-being and QOL are used interchangeably throughout the thesis. 3

Introduction

1.3 Thesis outlines

This thesis is organized in nine succeeding chapters. Details of the chapter description are discussed below.

Following the introduction chapter, Chapter 2 summarizes the contemporary literature on QOL and subjective well-being. It gives a general overview regarding the concept, approaches and theories of QOL. Moreover, the QOUL concept and its relationship with urban characteristics and person-place-activity approach are also discussed in more details. Identifying the main gaps in the literature provides a platform for constructing the theoretical framework, which is outlined in this chapter.

The data collection procedure and sample characteristics are described in Chapter 3. Based on our theoretical framework, a questionnaire was designed. The structure of questionnaire is divided into six modules, starting from socio-demographic information, and then the four main urban life domains (housing, neighborhood, transport and job), which are followed by the last module containing other life domains and overall QOL questions. After explaining questionnaire design, this chapter discussed the selected areas for our empirical research. Finally, a descriptive statistics for the collected data is reported.

Chapter 4 explores the first domain of urban life, and thus focuses on housing well-being. This chapter begins with reviewing the housing literature and the main theories and factors influencing housing satisfaction are discussed in detail. Based on the insights from the literature, and the link among various house attributes, the conceptual framework of this chapter is elaborated. A regression model is first used to analysis data, which is followed by estimation of a path analysis model. It is then followed by presenting the results, discussion and analyzes of the results and conclusion.

Chapter 5 studies neighborhood well-being domain, and starts with reviewing the main literature, concepts and approaches including the identification of main sub- domains and factors affecting neighborhood satisfaction. A conceptual model for this domain is designed to investigate the role of different attributes on neighborhood well- being. After using regression analysis, a structural equation model was built to estimate the relationships among multiple unobserved constructs (latent variables), and observable variables. This chapter also discusses and interprets the results, and comes with a set of conclusions regarding factors affecting neighborhood well-being.

Introduction

Following our domain-specific satisfaction approach, after housing and neighborhood, transport well-being is presented in Chapter 6. To achieve the objectives of this chapter a theoretical model is elaborated based on an extensive literature study, and understanding the influence of various factors on travel satisfaction. In this chapter, the SEM method is used to analyze the complex relationships between various objective and subjective indicators. Then, the results are discussed to investigate different factors affecting transport well-being and the important ones are presented in conclusion as the main factors correspond to transport well-being.

Chapter 7 offers detailed information on work well-being. It first introduces the concepts, theories and main factor influencing job satisfaction. Then, it describes the theoretical framework, which is estimated using regression analysis and path analysis method. It is followed by presenting and interpreting the results. It ends with a discussion and some important conclusions.

Chapter 8 turns towards the estimation of the effects of multiple life domains on the overall QOUL. This chapter starts with literature review of subjective well-being and the influence of major life domains on QOUL, and on each other. Then, the theoretical model is presented and the first estimation is done by regression analysis, which is followed by path analysis method. The results of both methods are presented and discussed, and important domains affecting QOUL are highlighted. This chapter presents a good framework for QOUL estimation and the main factors affecting it.

Finally, chapter 9 concludes the thesis. It summarizes the main findings of this research project, discusses the implications and limitations of the research and delineates directions for future research. This chapter shortly review all of the important conclusions derived throughout this research from domain-specific approach, and illustrates how different aspects of different life domains can affect overall QOUL. It also indicates to what extent these results are applicable.

Introduction

Concept and literature review

2.1 Introduction

For many years, social scientists have been examining Quality of Life (QOL) using aggregate data and focusing on aspects such as crime and economic challenges. In parallel, psychologists have been working on QOL as a subjective phenomenon with affective (emotions) and cognitive (knowledge) dimensions, which are deeply rooted in personality traits of individuals (e.g. Diener, 1994). During the last decade, urban planning scientists have also become interested in this concept, and the QOL topic has re-emerged with a new focus on urban settings as a response to challenges that cities are facing. Particularly, urban growth and migration have been acknowledged as the consequences of QOL differences among places. These differences concern both objective features of living places and people’s subjective assessments of Quality of Urban Life (QOUL). Therefore, in order to understand QOUL, we need to investigate

Concept and literature review how residents judge urban places, and use them to conduct their daily activities. Clearly, individuals have different judgments regarding attributes of urban life, which influence their QOUL. To address these aspects in a meaningful theoretical framework, it is essential to provide an overview of the theoretical and methodological evolution of the concept.

This chapter is organized as follows. First, we will discuss the notion of QOL and the evolution of the concept and measurement methods. Next, we will provide an overview of the QOUL concept and related theories namely, domain-based and person- place-activity. This will be followed by presenting the conceptual framework underlying this dissertation.

2.2 QOL (well–being)

2.2.1 Background

The history of the QOL debate is quite long and can be traced back to the beginning of the intellectual era when Greek philosophers debated the nature of good life or happy life (Sirgy et al., 2012). During the history, specifically the last century, many researchers worked on happiness. However, the concept of subjective well-being belongs to the contemporary era, and has been first captured by Ed Diener (1984), when he conducted empirical research based on a survey. Since then, many researchers elaborated the mainstream process of understanding, defining and examining subjective well-being in various disciplines. Despite the popularity of the topic, there is no consensus among researchers in terms and definitions of the concept.

When it comes to the notion of quality of life, complexity and multidimensionality overshadow the whole concept, as it is an all-embracing concept. Finding a precise definition for QOL is challenging as both terms “quality” and “life” reflect uncertain perspectives and paradigms. “Quality” in general is a characteristic or attribute regarding the degree of fineness and is perceptual and conditional. The American society of quality defines it as: "A combination of quantitative and qualitative perspectives for which each person has his or her own definition”. “Life” itself is also a difficult phenomenon to define. It is basically a characteristic and process that shows existence, growth, and functional activity and contains all sorts of senses and beliefs. Therefore, it is not surprising that the combinational term QOL is even more ambiguous and difficult to define. Pinto et al. (2017) argued that the roots of the QOL concept in

Concept and literature review sociology, geography, economics and health reflect the broad variety of dimensions, terminologies and definitions that are linked with the concept. Among those, “Happiness”, “Life satisfaction”, “Livability” and “Well-being” are the main notions that address the theme in the academic literature. “Happiness” is the state of emotions, derived from experiences and life events, and as Kahneman et al. (1999) stated it is associated with real-time moods and feelings. “Life satisfaction” is the outcome of evaluation and comparison of current life with ideal life, so it is the experienced life versus the expected life (Diener et al., 2013). Furthermore, “Livability” refers to the objective quality of life and is usually used for city rankings and material comfort in terms of goods and services (Okulicz-Kozaryn, 2013). “Well-being” is the most frequent term used interchangeably with QOL and as Pinto et al. (2017) advocated it reflects both emotions and the degree of functionality in a person’s life. All in all, considering the QOL notion and its objective and subjective implications, both “Life satisfaction” and “Well-being” can be used to explain the construct. Fortunately, the lack of consensus has not prevented empirical progress in the field. In the following, we will discuss the main theories and approaches associated with the concept of QOL.

2.2.2 Theory and measurement Over time, researchers achieved considerable progress in developing methodological and theoretical approaches concerning QOL, and thus the field has witnessed new perspectives and paradigms. Such advances have made QOL a field in transition, which motivates researchers to go beyond the accepted methods and examine new domains (Glatzer et al., 2015). As QOL studies vary based on the scale of the study, time and context of the research, there is no universal method to measure QOL. However, two principal approaches, hedonic and eudaimonic, contribute to QOL theoretically and methodologically. The hedonic approach focuses on happiness and thus associates QOL with pleasure, while the eudaimonic approach mainly addresses meaning, self- actualization and life purpose (Sirgy, 2012). These two views have been used in many studies as they both contain important aspects of well-being. Some studies used mixed methods and found that some domains of life contribute to both hedonic and eudaimonic well-being as they enable people to experience both happiness and meaningfulness. For instance, Delle Fave et al. (2011) found that family life and social relations contributes to both happiness and meaningfulness. They finally concluded that it is difficult to find clear borders in implications of life domains. Table 2.1 presents some of the most important theories on well-being that are mentioned in “The 9

Concept and literature review psychology of quality of life” (Sirgy, 2012).

Table 2.1 Integrative theories of QOL

Integrative Representative Theories of Main characteristics Conceptualization authors QOL

Life satisfaction is strongly Societal resources, influenced by life Livability Personal resources, experiences and is the Veenhoven (1995) theory Individual abilities. consequence of pleasures and pains of life events.

A long and healthy life, Capabilities determine Access to knowledge, Capability what people can do to A decent standard of Sen (1993) Theory achieve their potential living.

QOL consists of individual’s subjective and Personal and objective sets of

psychological circumstances. So both Quality of the attributes, psychological Lane (1994) Person+ Environmental characteristics of a person Environment conditions. and environmental conditions determine well- being.

Life satisfaction is Lance et al. (1995), Spillover Bottom-up effect of determined by satisfaction Brief et al. (1993) Model variety of life domains. from different life domains.

QOL is viewed as involving Short-term happiness, three constructs: QOL, Long-term happiness, happiness (short term& Argyle (1996) happiness Life Satisfaction, long term) as the affective component, life Absence of Ill-Being. satisfaction as a cognitive

Concept and literature review

component and the absence of ill-being.

The anticipation stage, Well-being The planning stage, conceptualization is dynamic and contains Dynamic The behavior stage, Dolan and White different stages. Focusing Well-Being The outcome stage, (2006) at different stages of well- The experience stage, being process give The evaluation stage. different results.

Unifying the affective and

Ontological Past well-being, cognitive dimensions Well-Being along with temporal Present well-being, Şimsek (2009) and the 3P perspective: past, present Model Future well-being. and future in a 2 × 3 matrix

2.2.3 Bottom-up spillover theory

One of the main theories that has been applied in many empirical QOL studies is bottom-up spillover theory (e.g. Woo et al., 2016). The theory conceptualizes the relationships between life domains and overall QOL in a hierarchy structure and focuses on present stage of life. In other words, life satisfaction assumed as a hierarchy with overall QOL at the top of hierarchy, domains satisfaction in the middle of hierarchy, and life events and broad situational and socio-demographic factors at the bottom of hierarchy. This theory is also very consistent with need hierarchy theory, which the fulfillment of the basic human needs can lead to happiness (Sirgy, 2012). Despite the popularity of the approach, there is no agreement among researchers in terms of life domains, and a broad variety of life domains have been included in QOL studies. Table 2.2 summarizes the life domains that have been addressed in some empirical studies.

2.2.4 QOL in environmental psychology: the concept of “person- place-activity”

When addressing human QOL, it is essential to consider the role of place. The tie between humans and the built environment has become the fundamental basis for a relatively recent interdisciplinary field called environmental psychology (Proshansky et

Concept and literature review

Table 2.2 Life domains in QOL studies

Main life domains in QOL studies

Author Financial Health Family Social Leisure Job

Carlquist et al. (2017) * * * * * *

Pietersma et al. * * * * * * ( 2014)

Loewe et al. (2014) * * * * * *

Dolnicar et al. (2012) * * * * * *

Sirgy et al. (2010) * * * * * *

Costa (2008) * * *

Praag et al. (2003) * * * *

Cummins et al. * * * (2003)

Vestling et al. (2003) * * * * * *

al., 1970). Environmental psychologists are interested in the study of human- environment interaction and predicting the conditions under which humans may behave well. In other words, they conceptualize QOL in the relation with the capacity of physical environment to fulfill human needs.

For many years, environmental psychologists have emphasized the fundamental role of place in defining the individual’s self-identity (e.g. Bernardo & Palma-Oliveira, 2016). As identity relates to concepts such as human territory and belonging, the link between living places and perception of identity has attracted attention. As a result, researchers found a strong association between sense of place and favorable assessments of place. Particularly, the key role of residential places in identity shaping and sense of attachment has been widely discussed by epistemologists. For instance, Clark et al. (2017) asserted that place identification strongly contributes to residential satisfaction. Later, other studies extended the perspective by defining different geographical scales in shaping identity and focused on various spatial units in perceived quality of living environments. This led to various studies examining the link between neighborhood and community in obtaining sense of satisfaction. Soon after, 12

Concept and literature review the functionality of place and the dynamic human life in time and space broadened the scope of the studies from static to more dynamic. Following that, researchers started assessing places not only in terms of the identity they make, but also by their functions and capacities of hosting different activities. In other words, daily life experiences of individuals were assumed to be broadly influenced by activities they commit in certain time and place. As such, work places have received a lot of attention and satisfaction with job as an important life domain has been included in well-being studies. More recently, mobility has broadened the perspective and some studies have focused on the link between (dis)satisfaction one gets from mobility and overall well-being. Activities such as social and leisure activities have also been taken into account.

This new perspective is also in line with both the hedonic and eudaimonic approach in well-being studies. Baumeister and Forgas (2018) argued that both hedonic and eudaimonic well-being can be achieved through engaging in activities that fulfill human emotions and potentials. Moreover, Steger et al. (2013) used the term “eudaimonic activities” and examined the link between eudaimonic daily activities such as volunteer work or participating in a meaningful conversation and well-being. They concluded that the more individuals participate in eudaimonic activities, the greater the well-being.

2.3 Quality of Urban Life (QOUL)

Currently, most of the world population is living in cities and metropolitan areas. Therefore, concern over the life quality in modern cities is the major reason that scientists invented the term QOUL to discuss issues regarding contemporary life in modern societies. A fundamental assumption underlying many QOUL studies is that characteristics of the urban environment and physical conditions may influence the degree of life satisfaction that residents experience. Marans and Stimson (2011), defined UQOL as: “satisfaction that a person receives from surrounding human and physical conditions, conditions that are scale-dependent and can affect the behavior of individual people, groups such as households and economic units such as firms”.

Despite the methodological advances in QOL studies by social scientists, it took long time for urban planners to shift from traditional approaches that were mainly focused on objective measure (such as gross national income) to subjective evaluations of urban features. Although, there is still no agreement on how best to measure QOUL, most of the researchers have accepted the importance of subjective evaluation of different living

Concept and literature review aspects (Tay et al., 2015). Considering the subjective evaluations in estimating QOUL, urban scientists tried to adapt theories from various disciplines such as economics, sociology and psychology. Domain–specific approaches are probably most popular in estimating QOUL that has been borrowed from psychology (e.g. Bloch-Jorgensen et al., 2018). The method assumes that QOUL is a function of evaluations made in different life domains, which are the results of objective characteristics of the environment or situational factors (e.g. socio-demographic factors). Given that the underlying dimensions of QOUL are numerous, different studies have selected different domains, sub-domains and attributes to investigate QOUL. Table 2.3 presents some of these studies. However, as explained earlier, using the concept of person-place-activity from environmental psychology can provide meaningful theoretical base for determining the relevant domains in QOUL studies.

2.4 Conceptual framework

As stated above, the domain-specific approach based on bottom-up spillover theory can reflect the nature of urban settings in explaining subjective well-being. Therefore, the assumption that physical characteristic of the environment may potentially influence the subjective assessment of the relevant domain, which in turn influences QOUL, can perfectly be tested by this method. Moreover, overall QOUL is hypothesized to be a function of the person-place-activity relationship, and thereby the main life domains can be extracted from the concept. Regarding place, housing well-being and neighborhood well-being have been included in the model, as both house and neighborhood are key living places. In terms of main urban activities, work well-being and transport well-being have been added to the model as both may significantly influence QOUL (Fig 2.1).

It is noteworthy to mention that these domains have been chosen due to their direct link to the built environment, considering the fact that urban life concept is very broad and it is almost impossible to include all possible domains within a single study. After conceptualizing urban life, it is time to include other important life domains that may impact QOUL. In relation to person notion, health, family and income are the three determinant factors. Therefore, three domains of health well-being, family well-being and income well-being have been added to the model. Furthermore, satisfaction with social activities and leisure activities can have a huge impact on subjective well-being. Thus, these two domains are also included in the model. In total, nine domains will

Concept and literature review

Figure 2.1 Life domains in relation to person-place- activity concept

explain QOUL and among those four domains are hypothesized as urban life domains and will be discussed in separate chapters. The other five domains will be included in the final measurement of QOUL. Moreover, all these domains may have effects on each other as satisfaction in one life domain can spill over to other life domains. Thus, the links among domains are also added to the theoretical framework. Socio-demographic information is also considered in this framework, and used as a predictor of well-being in life domains and also overall QOUL (Fig 2.2).

Accordingly, the conceptual framework is concerned with the following concepts:

1. Detail-oriented study of QOUL by focusing on four domains of:

 Housing well-being

 Neighborhood well-being

 Transport well-being

 Job well-being

2. Overall QOUL consisting of nine major life domains.

Concept and literature review

Table 2.3 Summary of recent QOUL studies

Study Author(s)/ Aspects of QOUL studies

country (Domains/indicators)

Personal and Subjective assessments of 11 QOL domains: Overall Neighborhood standard of living, , housing, family life, social life, Indicators of financial life, health, leisure activities, employment Quality of Urban status, freedom /independence, amount of time for Life: A Case Study Low et al. yourself, amount of money for yourself. And, of Hong Kong (2018)/ China subjective assessments of attributes in 3 levels of living domains: Housing (both subjective and objective indicators), neighborhood and city as a whole. Socio-structural relations: Satisfaction with: safety feeling at city, safety Smart city and feeling at neighborhood, state of the streets and quality of life: buildings Citizens perception Environmental well-being: Macke et al. in a Brazilian case Satisfaction with: green space, cleanness, public (2018)/ Brazil study space, cultural facilities, air quality, presence of foreigners Material well-being: Satisfaction with: life, financial situation, place I live Community integration: People of city are trustable, people of my neighborhood are trustable, public administration is trustable Urbanism and Okulicz-Kozaryn City characteristics: crime, housing stress, low happiness: A test & Mazelis education, low employment, persistent poverty, of Wirth’s theory (2018)/ United population loss, personal income (USD 1000)/cap, of urban life states and race. And, subjective satisfaction with life

From Spontaneous Subjective Evaluation of: to Planned Urban Education, job, living standard, health, Development and Peiser (2016)/ accommodation, family relation, social life, financial Quality of Life: the Vietnam condition, housing quality, Case of Ho Chi pollution/noise/congestion, adequacy of park, tree, Minh City road, school, hospital, and recreation places, quality of basic places, and using public facilities, social complexity, the interaction with neighbors and overall satisfaction.

Concept and literature review

Mediating Effects Between Objective and Subjective Von Wirth et al. Satisfaction with: Region, city, neighborhood Indicators of Urban (2015)/ Quality of Life: Switzerland Subjective assessments of safety indicators and Testing Specific access indicators. Objective access and objective Models for Safety threats for safety were also measured by related and Access attributes.

Physical domain: neighborhood, house, natural environment, garbage collection, public transport, street condition, water quality, traffic congestion and Evaluating Quality calmness. of Life in Urban Rezvani et al. Economic domain: employment opportunities, own Areas (Case Study: (2013)/ Iran Noorabad City, economic condition, distribution of wealth,

Iran) educational facilities, health care facilities and recreational facilities. Social domain: sense of personal security, health conditions, happiness, sense of belonging, relations with neighbors, reliability of inhabitants, hope for the future, success in life.

An evaluation of

residents‟ quality Social life, good condition for children, safety, Sedaghatnia et of life through greenery and quietness, transport and community al. (2013)/ neighborhood facilities and services such as shopping centers, Malaysia satisfaction in leisure facilities, schools, universities. Malaysia The Quality of Life Marans & Three domain of house, neighborhood and in Metro Detroit at Kweon (2011)/ community Indicators: Family, Health, Friends, job, the Beginning of United States Standard of living, things want to do, Leisure time. the Millennium Looking at three levels: dwelling, the micro Measuring Quality Türkoğlu et al. neighborhood and the macro neighborhood of Urban Life in (2011)/ Turkey Other domains: family life, health, job, friends, Istanbul standard of living, leisure activities, and satisfaction with life as a whole.

Subjective Quality of Life in McCrea et al. Four main attributes of urban environments: access Queensland: (2011)/ to services and facilities, noise pollution, incivilities Comparing Australia and social capital. Metropolitan, Regional and Rural Areas

Concept and literature review

The Brisbane- South East Queensland Stimson et al. Overall transportation, Economic conditions, Health Region, Australia: (2011)/ services, Social conditions, Educational services, Subjective Australia Natural environment, General services & facilities and Assessment of Lifestyle. Quality of Urban Life and Changes over Time

Years of residence at the address, public transport quality, distance to public transport, housing The Salzburg Keul & Prinz Quality of Urban satisfaction, food shops and food shop quality, use of (2011)/ Austria Life Study with GIS green space, leisure space accessibility, Support neighborhood quality, assessment of safety and threats.

A Perception

Survey for the

Evaluation of

Urban Quality of Education facilities, Quality of environment, Safety, Life in Kocaeli and Senlier et al. Public transport, Neighborhood Social and cultural a Comparison of (2009)/ Turkey the Life facilities, Sufficiency of health services, Quality of Satisfaction with health services. the European Cities

The Quality of QOL domains:

Urban Life and Satisfaction with: family life and friends, health, Neighborhood job/school, standard of living, and life as a whole. Oktay Satisfaction in & Rustemli QOUL domains: Famagusta, (2011)/ Cyprus satisfaction with: the individual home or dwelling, Northern Cyprus the immediate (microscale) neighborhood, and the overall (macroscale) neighborhood.

Health, Climate, Crowding, Sporting, Housing

conditions, Travel to work, Environmental pollution Possibilities and Türksever & Shopping facilities, Education provision, Cost of Limitations for the Atalik (2001)/ living, Noise levels, Job opportunities, Relation with Measurement of Turkey the Quality of Life neighbors, Parks and green areas, Leisure in Urban Areas opportunities, Crime rate, Accessibility to public transportation, Traffic congestion.

Concept and literature review

A Method for Community, Family, Finances, Personal health, Assessing Leisure life, Spiritual life, Job, Education, Friendship, Residents' Sirgy et al. Neighborhood, Environment, Housing, Cultural life Satisfaction with (2000)/ and Social Status. Community-Based United States Services: A QOL Perspective

To briefly explain the four main urban life domains, the first is housing well- being, which is defined by satisfaction with dwelling attributes. These subjective evaluations are hypothesized to be influenced by various physical characteristics of the house. The second domain is neighborhood well-being with many attributes categorized in three sub-domains, namely satisfaction with: neighborhood characteristics, neighborhood physical activity and neighborhood services. Moreover, transport well- being is conceptually related to satisfaction with trip characteristics and built environment characteristics. Trip characteristics consist of trip attributes, depend on mode use and time frame of the trip. Built environment characteristics consist of the attributes of neighborhood satisfaction and accessibility satisfaction.

Finally, job well-being or work satisfaction domain is defined by well-being in three sub domains: work conditions, workspace and work location. Each sub-domain contains relevant attributes that reflect the satisfaction with certain job characteristics.

2.5 Conclusions

Despite the considerable amount of QOL literature particularly in social science, the concept is still surrounded with ambiguity and lack of consensus in both theory and methodological approaches. This chapter, however, has attempted to provide an overview of the dominant studies of QOL and QOUL, and to clarify these concepts and theoretical background. Furthermore, the insights from this systematic literature review provide the theoretical basis for the empirical analysis in this thesis. The summary of the literature review has shown that the vast majority of QOUL studies have examined well-being in different life domains using sets of indicators, and conceptualized QOUL as the sum of satisfaction with different life domains. However, these studies have not examined urban life domains in detail, as it requires detailed information on residents’ daily living and also separate analysis for each domain. This provides the motivation for conducting research on QOUL to explore the extent to which urban life domains and

Concept and literature review

Figure 2.2 The conceptual framework

built environment factors can influence overall QOUL of individuals. To adequately explore this question, it is essential to use the theoretical frameworks, and collect data to operationalize the frameworks within a particular context. The following chapters describe the data collection and analytical findings based on the concept.

Data collection and descriptive statistics

3.1 Introduction

The main objective of this study is to investigate the influence of built environment and urban life factors on overall QOUL and subjective well-being of residents. As discussed in Chapter 2, the four domains, viz. housing well-being, neighborhood well-being, transport well-being and job well-being are defined as the key urban life domains related to the built environment, so they will be discussed in separate chapters of this dissertation. In order to analysis the effects of different attributes on these domains and also overall QOUL, detailed information regarding each domain is required. The best way to obtain this information is to collect data by designing a survey. In this chapter, first the survey design will be systematically explained for each QOUL domain. After discussing the selection of study areas, sample selection and statistical properties of the samples will be reported.

Data collection and descriptive statistics

3.2 Survey design

In this section, we will summarize all the components of survey design in a step-by-step manner as follows:

3.2.1 Data consideration

The data requirements for this study can be segmented into six modules: Socio- demographic information; Housing well-being module; Neighborhood well-being module; Transport well-being module; Job well-being module and other life domains well-being (family life, social life, leisure life, financial life and health) as well as overall QOUL module. Before each section of survey, respondents were informed about the objective of the questions as they were prompted to answer to detailed questions.

3.2.2 Survey instrument

The survey digitalized in a system called “Berg system” developed by DDSS unit of TU/e. As the survey is quite long and demanding, and has aimed to focus on specific neighborhoods in the city of Eindhoven, it was not sufficient to approach respondents via an opt-in Panel or social media platforms (such as Buurtpreventie group). Therefore, the door-to-door approach was used in which residents were handed a flyer (in both Dutch and English) with an explanation of the research goals and the webpage of the questionnaire (Fig. 3.1).

The data were collected in the summer and autumn of 2016 (from July to November). Data were collected during different hours across multiple days of the week, so that both part time and full time working people could be contacted. To clarify the general purpose of the study and reduce any misunderstanding, a Dutch student cooperated and explained the survey to residents. The questionnaire was designed into separate parts to reduce halo effects. The response rate was 31% (222 out of 720).

3.2.3 The questionnaire

In general, the online survey was designed in a way to prevent most of the commonly made mistakes. Therefore, except age and postal code, all other questions are closed ended questions, ranging from multiple-choice questions to rating scale questions. Also, in order to avoid missing data as much as possible, respondents had to answer the questions to be able to proceed to the next page of the survey. Satisfaction was measured using a 10-point Likert scale.

Data collection and descriptive statistics

Figure 3.1 Flyer of the survey As mentioned earlier, the questionnaire consists of six modules (Appendix 1 presents the full questionnaire). On the first page, respondents were informed about the general aim of the study. The following modules were included in the indicated order:

The first module is related to socio-demographic and general information. In this module, respondents were asked to provide information regarding their age, gender, education, marital status, nationality, employment status, household size, length of

Table 3.1 Number of respondents per area in data collection approach

StrijpS Schoot Villapark Doornakkers

Total Number 55 53 55 59 of respondents

Number of 27 38 55 52 door by door respondents

Number of 28 15 0 7 social media respondents

Data collection and descriptive statistics residency, car ownership and also the postal code of the household. However, household income was asked separately among the last questions of the survey and apart from the options of income amount, the option “I don’t want to say” was also included as income might be considered confidential for some of the respondents.

The second module concerned housing well-being. This module included questions regarding the following attributes:

1. Ownership status: residents were asked whether they own or rent their current property.

2. Housing cost: respondents were asked to clarify the monthly price they pay for their property containing six options, viz. zero euro per month (in case the households fully owned the property), 500 to 750 euro, 750 to 1000 euro, 1000 to 1250, 1250 to 1500, and 1500 and higher.

3. Dwelling characteristics: respondents were requested to clarify the following characteristics of their dwelling:

 Type of dwelling, containing four options: flat house, semi attached house, attached house and villa house.

 Age of dwelling, containing ten options: after 2010, 2001-2010, 1991-2000, 1981-1990, 1971-1980, 1960-1970, 1945-1959, 1931- 1944, 1906-1930, before 1906.

 Size of dwelling, containing three options: less than 60 m2, between 60m2 and100 m2 and more than 100 m2.

 Design of dwelling containing two options “It fits my lifestyle” and “It does not fit my lifestyle”.

 Exterior space of dwelling, containing seven options: balcony, garden, roof terrace, roof terrace and balcony, roof terrace and garden, balcony and garden, without any exterior space.

 Renovation condition of dwelling, containing seven options: no need for major repair, kitchen needs repair, roof needs repair, floor needs repair, bathroom needs repair, piping or wiring system need repair, different parts of house need repair.

 Number of bedrooms, containing four options: No bedroom, one bedroom, two bedrooms, three bedrooms and more.

Data collection and descriptive statistics

 Safety/security facilities of dwelling, containing three options: no special facility for security, traditional facilities (Fence, etc.) and high- tech facilities (CCTV, sensors, etc.).

 View of the house, containing six options: green area, blocks of houses, street with shops, highway, vacant space and other.

 Parking space of house, containing two options: free parking and paid parking.

4. Satisfaction with ownership status, housing cost and all the dwelling characteristics. Respondents were asked to rate their degree of satisfaction regarding all the aspects of their house on a 10-point scale, ranging from 1 (not satisfied at all) to 10 (extremely satisfied).

5. Finally, respondent were asked to evaluate their overall satisfaction regarding their house on a 10-point scale.

The third is related to neighborhood well-being. As subjective evaluations have been shown to be more important in predicting neighborhood satisfaction (e.g. Parkes et al., 2002; Oh, 2003), in this module, respondents were asked to rate their degree of satisfaction (from 1 to 10) with various neighborhood aspects. However, these aspects have been categorized into four main segments:

1. Satisfaction with neighborhood characteristics consists of the questions regarding satisfaction with: safety, cleanness, greenery/landscape, noise pollution, water quality, road/pedestrian quality, lighting, traffic congestion, neighborhood reputation, personal attachment, livability and housing upkeep and architecture. At the end of this part respondents were asked to rate their overall satisfaction regarding neighborhood characteristics.

2. Satisfaction with neighborhood physical activity (walking/biking condition). For walking, satisfaction with traffic safety, quality of path, safety regarding crime and air quality were asked. For biking, satisfaction with: traffic safety, quality of path, size of road and greenery of route were asked. At the end of this part, respondents were asked to rate their overall satisfaction regarding neighborhood physical activity support.

3. Satisfaction with neighborhood services consists of evaluating six major destinations at neighborhood level. These destinations include grocery

Data collection and descriptive statistics

shopping, sport centers, cafe/restaurants, medical centers, kids' school/daycare, park and other shopping stores which were asked to be evaluated by travel time, time schedule, quality and crowdedness. At the end of this part respondents were asked to rate their overall satisfaction regarding neighborhood services.

4. Finally, respondents were asked to evaluate their overall satisfaction regarding their neighborhood. In order to make this evaluation easier, respondents were briefly informed how they have already evaluated previous questions. This can be seen in Fig. 3.2.

Forth module is transport well-being. This module contains two general groups of questions:

1. The questions regarding travel characteristics, which consist of time frame and mode use including the travel attributes evaluation. So basically, respondents were asked about the transport mode of their daily trips in both peak and off-peak hours, and should have chosen among the five options viz. “bike”, “walk”, “car use”, “public transport use”, “both public and private modes” and “I don’t travel”. If the respondents chose “car”, “public transport” or “both public transport and car” in the previous question, then they were guided to the next page for evaluating their trip attributes. Car commuters were asked to evaluate their satisfaction with: travel time, travel cost, travel comfort, travel safety, traffic congestion, parking availability and parking cost. While, public transport users were asked to evaluate seven aspects, viz. satisfaction with: travel time, travel cost, travel safety, waiting time, stop/station distance, crowdedness, inside transport means quality.

Questions regarding built environment characteristics, which potentially can influence travel satisfaction. In this regard, respondents were asked to rate their satisfaction level with travel time to the major places, viz. work place, kids school/day care, family/friends, service places and recreational centers. More questions regarding physical environment have already been addressed in neighborhood well-being module and can be recalled for the analysis.

Data collection and descriptive statistics

Figure 3.2 An example of the final page of neighborhood module The fifth module asked about job well-being. In this module, first, respondents were asked if they are employed or not. If they answered they were unemployed, they automatically skipped all the questions of this module. Also, those who were employed should clarify their employment type (full-time vs part-time). In this module three categories were used to measure subjective work satisfaction as follow:

1. Work conditions satisfaction where respondents were requested to rate their satisfaction with: Workload, quality of work, salary and people at work and finally the overall satisfaction with work conditions.

2. Work location satisfaction where respondents were asked to evaluate their satisfaction with travel time to work, availability of public transport at work area, safety of work area, access to shops at work area, and finally the overall satisfaction with work location.

3. Workspace satisfaction where respondents were asked to evaluate their satisfaction with size of workspace, access to office, privacy, noise level, and furniture/tools, and finally the overall satisfaction with workspace.

The sixth module is about other life domains and overall QOUL. In this module, respondents were asked to evaluate the degree of their satisfaction regarding five key life domains, viz. family life, health well-being, social life, leisure life and financial life. Moreover, they were asked to mention if they recently had lost a person in their family. Finally, in the last page of survey respondents were requested to evaluate their overall QOUL as follows: “Considering everything in your life, how do you rate the overall quality of your life?”

Data collection and descriptive statistics

3.3 Study areas

In order to understand the role of urban living experience and built environment characteristics on micro-level subjective well-being, sufficient variability in neighborhoods is needed. Therefore, we collected data in four different areas in the city of Eindhoven. Two of these areas, namely Villapark and Doornakkers are located in East Eindhoven in the Tongelre district and the other two namely, StrijpS and Schoot in the Western part of the city in the Oud-Strijp district (Fig. 3.3). These four areas are totally different in terms of their socio-economic position, physical characteristics, dwellings type and design (Fig. 3.4). StrijpS is a former industrial area, which is now becoming a unique smart community and innovation center with strong impact on the city. About 94% of houses are rented and all residential houses are apartments, and were built after 2000. Residents in StriipS are mostly young and educated, and combine working and living in apartments and studios. Compared to StrijpS, Schoot has a longer residential history, and combines old and new houses with various characteristics and designs as is belonged to different architecture eras. Property types are also diverse, ranging from apartments to semi attached and attached houses and villas. However, 67 % of the properties are rented. Villapark is among the most affluent areas of Eindhoven with 78% native Dutch residents. Most houses were built before 1945 and about 68% are owner-occupied, while just 23% are apartments. In contrast, Doornakkers is listed among the 40 most deprived neighborhoods of the Netherlands because of its socio- economic problems. It has the lowest average household income and highest unemployment rate among our study areas. Most of houses in Doornakkers were built after the second world war for working class residents, and are mostly attached or apartment buildings. In general and according to neighborhood thermometer of Eindhoven2, Villapark area is ranked above average, StrijpS and Schoot are average areas and Doornakkers is below average. In addition to fundamental differences in both the physical environment and socio-economic position, these areas have clear geographical boundaries, which make a clear “socio-spatial schema” for residents to evaluate their neighborhoods. Table 3.2 shows the general characteristics of the neighborhoods and their inhabitants (based on data from Statistics Netherlands (CBS), 2018).

2 http://www.buurtthermometer.nl/ 28

Data collection and descriptive statistics

Figure 3.3 Location of the four neighborhoods in Eindhoven 1: StrijpS, 2: Schoot, 3: Villapark, 4: Doornakkers

3.4 Sample selection and characteristics

Of the 720 contacted households in the four areas of Eindhoven, in total 222 (31%) completed the questionnaire. The questionnaire took between 30 minutes and one hour to complete. To obtain a high quality dataset, a data checking and cleaning process was carried out. This process consists of different stages: identifying invalid cases (statistical outliers), and handling missing data. In the first step, cases with invalid postal codes were found and eliminated (8 cases). In the second step, cross data checks were employed to enhance data accuracy. That is, different answers were matched with each other to detect contradictions. For instance, those respondents who chose the same satisfaction rating for many questions were recognized as inconsistent. Therefore, another 5 cases were removed from the original dataset. After removing 13 cases, eventually, 209 valid cases were chosen for our analyses. Tables 3.3 and 3.4 show the socio-demographic characteristics of the respondents and the descriptive statistics of domains satisfaction in the four areas, respectively. The descriptive statistics shows a larger share of female respondents (55.5%) compared to males (44.4%), and also a larger share of young (40%) and middle-aged respondents (46%) compared to seniors (14%). Regarding education, as seen in Table 3.3 higher education is heavily overrepresented in the sample. In terms of household income, as Figure 3.5 shows, the income level of respondents is quite representative of the areas’ population, since most respondents in Villapark are among the highest income households, while the proportion of low-income residents is the highest in Doornakkers area. 29

Data collection and descriptive statistics

Figure 3.4 Local dwelling streets of 1: StrijpS, 2: Schoot, 3: Villapark, 4: Doornakkers

Data collection and descriptive statistics

Table 3.2 General situation of the four neighborhoods

Area Number Owner- Rental Men/ Single Family Average

Neighborhood (km2) of Occupied houses Women /Two With income

1 to 4 houses houses adults kids (household)

StrijpS 0.30 1012 953 54 806/ 646/ 40 29.9 k

632 289

Schoot 0.39 1532 1208 504 1464/ 795/ 258 33.2 k

1263 357

Villapark 0.56 815 552 263 1089 239/ 263 58.5 k

/950 242

Doornakkers 1.22 2993 1153 1890 3332/ 1186/ 787 28.7

3059 742

Eindhoven 88.92 108.682 51225 57424 117.602 41.787 28.565 36.6 k 111.522 /29.425

Overall, the sample is not completely representative of the population of our study area. However, overrepresentation of women and higher educated people in the sample is common for most on-line surveys (e.g. Van den Berg, 2012). Moreover, as shown in Table 3.5, the majority of households concerns renters (57%), living in non- flat (59%) and small houses (57.5%). About 55% of the houses are old (more than 20 years old), and the majority has an exterior space (88%), and more than three bedrooms (52%).

A greater number of households (58%) pays a low price for their housing (less than 750 euro monthly), and 53% live less than 5 years in their current property. Most houses are in good condition and do not need any repair (77%), and a great portion of them is equipped with safety/security facilities (65%). In addition, people are generally the most satisfied with the type and age of their dwellings, and the least satisfied with the housing price, although among the car-owners households (78.0% of total) satisfaction with house parking has the lowest ratings (Table 3.6). Regarding neighborhood characteristics, air quality for walking and safety received the minimum average satisfaction, which could indicate some social and environmental issues in our study area (Table 3.7).

Data collection and descriptive statistics

Table 3.3 Description of social-demographic data (N=209)

Variable Category Number Frequency (%)

Gender Female 116 55.5

Male 93 44.5

Age Age 20-35 years 84 40

Age 36-64 years 96 46

Age>65 years 29 14

Living status Single 70 33.5

Non-single 139 66.5

Having kid (s) Household with kids 60 28.2

Household without kid(s) 149 72.8

Education Low educated 40 19

Higher educated (college degree) 168 81

Nationality Dutch 190 90.9

Non Dutch 20 9.1

Job Employed 130 62

Unemployed 79 38

Employment Part time 53 25

status Full time 77 37

Household size 1 person household 73 34.9

2 persons household 80 38.3

3 or more persons 56 26.9

Household below 2000 euro 62 29.7

income 2000-4000 euro 67 32

(11% missing) 4000 euro & higher 35 27.2

Car ownership No car 44 21.1

car owner 165 79.9

Data collection and descriptive statistics

Figure 3.5 Chart of household income in study areas

Regarding transportation, bike and motorbike are the most used transport modes during both peak and off-peak hours with around 36% of the commuters in peak hours, and 48% of the commuters during off peak hours. These commuters are also the most satisfied group of travelers (with a mean satisfaction of 8.17 for peak hours and 8.22 for off-peak hours). The car is the second most preferred transport mode (27.8%), and the number of car commuters during peak and off peak hours is quite close, as well as their mean satisfaction (7.68). Generally, during peak hours, bikers and non-travelers are the most satisfied commuters, but during off peak hours those who walk and bike are the most satisfied group. Table 3.8 lists the frequency of using different travel modes during both peak and off-peak hours and the mean satisfaction of commuters.

As most of the variables in housing section of survey are categorical variables with several categories, for further analysis, it is necessary to combine and merge various categories (with small frequencies) into a new category, which is presented in Table 3.6.

Data collection and descriptive statistics

Table 3.4 Descriptive statistics of life domains satisfaction

STRIJPS SCHOOT VILLAPARK DOORNAKKERS

Mean SD Mean SD Mean SD Mean SD

Housing 7.92 1.36 7.8 1.4 7.2 .8 7.6 1.3

Neighborhood 7.7 1.27 8 .9 7.96 .69 7.6 1.3

Job 7.6 .98 7.8 .86 7.82 .79 7.2 .92

Transport 7.6 1.0 7.1 .93 7.6 .92 7.4 1.04

Social life 7.7 1.09 7.6 1.2 8 .67 7.6 1.3

Health 7.7 1.04 7.7 1.4 8 1.2 7.48 1.84

Income 7.4 1.4 7.8 1.7 7.1 1.4 7.6 1.8

Family life 7.7 1.5 7.9 1.8 7.2 2.09 8 1.6

Leisure life 6.63 2.4 7.54 1.96 7.89 1.42 7,13 2.31

Overall QOUL 7.6 1.1 7.7 1.2 7.8 1.3 7.48 1.1

Table 3.5 Descriptive statistics of satisfaction with housing characteristics

rice

layout

Design/ security Safety/ P Monthly Parking

Ownership house of type house of size number Bedrooms space Exterior house of Age house of View situation Renovation

Mean 7.91 8.00 7.93 7.67 7.25 8.00 7.6 7.83 7.97 7.57 7.34 5.91 Satisfaction 8

Std. 1.54 1.53 1.57 1.79 2.23 1.61 1.7 1.63 1.54 1.31 1.74 3.73 Deviation

Data collection and descriptive statistics

Table 3.6 Description of housing data, and categories of categorical variables

Variables Original categories survey Number New categories after combining & recoding/ frequency (%)

Ownership Rent 120 Rental house/ 57%

Own 89 Owner occupied house/ 43%

Type of House Flat 86 Flat/ 41%

Attached 74 Non-flat/ 59%

Semi attached 35

Villa 14

Size of house Less than 60 48 Small house / 57.5%

60-100 72

More than 100 89 Large house/ 42.5%

Number of bedrooms Zero 30 Continuous variable

One 36

Two 34

Three and more 109

Construction year of After 2010 58 New house/ 45% House ( age of 2001-2010 36 house) 1991-2000 13 Old house/ 55%

1981-1990 8

1971-1980 3

1960-1970 6

1945-1959 30

1931-1944 11

1906-1930 41

Before 1906 3

Exterior Space Balcony 25 With exterior/ 88%

Garden 91

Roof terrace 30

Balcony and Roof terrace 2

Balcony and garden 18

Garden and roof terrace 18

No exterior space 25 No exterior/ 12%

Data collection and descriptive statistics

Housing price 0 Euro 23 Low price / 58%

500-750 Euro 97

750-1000 Euro 52 High price/ 42 %

1000-1250 Euro 14

1250-1500 Euro 12

1500 Euro & higher 11

Renovation situation No need for major repair 161 No need to repair/ 77%

Roof needs repair 7

Floor needs repair or change 3

Bathroom needs repair or change 11 Needs repair/ 23%

Kitchen needs repair 4

Piping /wiring system 1

Different parts need repair 22

Main view Green area 59 Green view/ 28%

Blocks of house 108 None Green view

Street with shops 10 / 72%

Highway 1

Vacant space 5

Other 26

Design/ layout It fits my lifestyle 184 Design fits/ 88%

It doesn't fit my lifestyle 25 Design not fits/ 12%

Safety/ security no special facility 78 No security/ 36.5%

some traditional security system 107 With security facility/ 62.5%

some high-tech security system 24

Length of residency Less than 1 year 23 Continuous variable

1 to 2 years 30

3 to 5 years 58

5 to 15 years 60

15 years or longer 38

Parking Free parking 111 Free parking / 53.1%

No free parking 98 Paid parking /46.9 %

Data collection and descriptive statistics

Table 3.7 Descriptive statistics of satisfaction with neighborhood characteristics

Std. Std.

Mean Deviation Mean Deviation

Noise pollution 6.59 2.017 Air quality 6.40 7.505

Traffic congestion 6.75 1.800 Architectural appearance 7.65 1.646

Air quality when walking 5.00 7.899 Personal attachment 7.36 1.819

Image/reputation 7.79 1.676 Cleanness 7.30 1.554

Quality of path for biking 6.30 3.280 Road/ pedestrian quality 7.63 1.533

Quality of shop (grocery) 7.78 2.481 Safety 7.29 1.406

Safety regarding crime for walking 4.82 10.724 Green space 7.23 1.753

Traffic safety for walking 4.79 10.719 Lighting 7.96 1.349

Traffic safety for biking 5.73 3.058 Livability 7.28 1.467

Road size for biking 6.30 3.211 Neighborhood services 7.79 1.015

Green route for biking 6.08 3.184 Neighborhood physical activity 7.35 1.337

Time schedule (grocery) 7.94 2.555 Neighborhood characteristics 7.59 1.218

Travel time (to grocery) 7.73 2.722 Overall neighborhood 7.83 1.103

Table 3.8 Descriptive statistics of satisfaction with transport characteristics

Mode of transport

Private Public Both Bike/ walk No trip (car) (Bus, public motor train, and bike tram) private

Frequency 27.8% 8.1% 4.3% 35.9% 4.3% 19.6%

Peak 7.67 7.88 7.44 8.17 7.33 8.17 hours Mean satisfaction

Frequency 28.7% 6.2% 8.6% 48.3% 6.7% 1.4% Off-peak

hours 7.68 7.15 7.67 8.22 8.43 6.63 Mean satisfaction

Data collection and descriptive statistics

Table 3.9 Descriptive statistics of satisfaction with work characteristics

Overall Work Work

work location conditions workspace

Employment Frequency satisfaction satisfaction satisfaction satisfaction

Mean 7.55 7.58 7.55 7.36

Part-time 41% Std. .972 1.379 .992 1.272 Deviation

Mean 7.70 7.62 7.66 7.56

Full-time 59% Std. .947 1.478 1.021 1.082 Deviation

Mean 7.64 7.61 7.62 7.48

Total 100% Std. .956 1.433 1.007 1.163 Deviation

3.5 Conclusions

In this chapter, based on our theoretical framework, the design and administration of the survey was presented. This survey aimed at collecting data on residents’ satisfaction regarding their living experience in urban neighborhoods and their overall life well- being. To capture the link between urban life characteristics and well-being, data were collected from four urban neighborhoods with completely different characteristics. Data collection was done on various days of the week in two seasons (summer and autumn 2016), so the weather influence on mood and subjective assessments was reduced to some extent. As the survey was long and demanding, the response rate was not high, and therefore, data collection took a long time. However, the high-quality data is well worth the effort. The high-quality data can be justified for several reasons: first, we had informed respondents upfront that the survey had been designed for academic research and was anonymous. Second, the survey website had a high quality surface and was easy to use (based on click and enter). In general, online surveys are more functional compared to interviews or paper-based surveys and needs less effort for respondents. Third, we conducted a pilot survey and based on the feedbacks both questions and the digital surface were improved. For instance, we created a responsive framework website, which means that the survey tables automatically changed to fit any device (such as mobile phone). 38

Housing well-being

4.1 Introduction

Adequate housing has always been acknowledged as a basic need, and an indispensable part of everyone’s life (e.g. Fernandez-Portero, 2016). Scholars, who studied the effects of different physical constructs upon quality of life, illustrated the potential insights that can be derived from housing research. Several empirical studies found a direct link between quality of life and housing satisfaction, which resulted in the recognition of housing as a criterion for determining quality of life (e.g. Burton et al., 2011; Costa-Font, 2013; Fernández-Carro, et al., 2015; Fernández-Portero et al., 2017; Zebardast & Nooraie, 2018; Zhang et al., 2018). On the other hand, whereas housing satisfaction research has received substantial attention in marketing, urban economics and sociology, it has only received marginal attention in the well-being research. This

Housing well-being little emphasis on housing in terms of quality of life measurement highlights the gap in this research stream and the need for more detailed studies on this interdisciplinary field.

In formulating the conceptual model of this PhD research, the concept of place plays a fundamental role, and helped capturing more insights on the relationship between human well-being and the surrounding environment. Here, housing will be considered as an important physical component of the built environment that affects residents’ satisfaction. Therefore, housing satisfaction contributes to quality of life as a main domain, and thus, this chapter will focus on it. Explicitly, the following research questions will be addressed:

I. What are the main predictors of housing satisfaction? II. To what extent do these indicators contribute to predicting housing satisfaction? III. What are the relationships among these indicators and to what extent do they affect each other?

4.2 Literature review

Over the past 50 years, many studies have been carried out to examine housing satisfaction. Comparing the current studies with the older ones shows that there have not been so many changes in the concepts, terminology and definitions associated with housing satisfaction. However reviewing these definitions, concepts etc. in other researchers’ work provides insights and a good platform for our conceptualization.

4.2.1 A place called house (terminology and definitions)

Human beings understand the world through their senses and the information they get from the environment, where their lives occur. In the epistemic relation between human and environment, they both influence each other and the existing interactive system among them. House in this regard is “a product of the society in which we live in” (Massey, 1995). This enables personal and interpersonal relationships by having fundamental impact on our perception toward life (Biswas-Diener & Diener, 2006).

“First we shape the place, then places shape us” (Winston Churchill), this statement has proven to be true, and house in this sense is not just where we live, but a sign that manifests who we are and how we live, and that’s why it has been an

Housing well-being evidence for civilization. In fact, a place, which labeled as a house is beyond a physical notion, but rather a source for stability, identification and recognition, which in turn should provide us, comfort, safety and a foundation for our spiritual life (Easthope, 2004; Vera-Toscano & Ateca-Amestoy, 2008). This implies that housing contributes to different level of needs at the hierarchy of needs (Maslow, 1954), started from a material structure responding to physiological needs and upgraded to higher level of motivational factors (Zavei & Jusan, 2012). In general, house, as a spatial unit in connection with its surrounding, has different functionalities. In its micro scale, it is a shelter that reflects living style and culture of its inhabitants. In the bigger scales, it can structure social life and be a demonstration for the attachments to the living environment (Jiboye, 2014; Aragonés et al., 2017). Therefore, it is an agent that intertwines between the personal and interpersonal environments and can be affected from both inside and outside. There is an interactive system between the way of living and the environment and this relation is a mutual influential one. House, in this regard, is the most substantial place that can affect perception toward life and can reflect social realities (Easthope, 2004). Staying in a place that fulfills most needs and aspirations is essential and since needs and aspirations change during the life cycle, housing satisfaction should be considered as a dynamic process (Ogu, 2002). Therefore, based on different stages of life, individual demand for housing services varies.

In summary, housing is as an inevitable demand and homeownership is a sign for proper financial situation. Having a house is always a big investment and an important financial decision since it is the most expensive object that someone can purchase in one’s lifetime (Flavin & Yamashita, 2002; Dusansky & Koc, 2007). Therefore, the conception of house is closely linked with the economy. Following utility theory, everyone wants to derive the maximum utility out of their house, while the various individual constraints lead to different levels of house satisfaction (Makinde, 2015). Moving for a better accommodation at different life stages of individuals brings the topic of mobility to housing research literature.

In general, housing research is a huge field beyond individual and household scale, and is also connected with regional, national and international issues such as migration, economic decline (recession) and other related topics. While there is an extensive amount of literature on housing, the majority of these studies is dedicated to housing displacements and mobility, and thereby, the focus has been on residential satisfaction as a combination of neighborhood and dwelling satisfaction. However, very

Housing well-being few studies have examined the link between housing satisfaction and well-being (Dekker et al., 2011; Li & Wu, 2013). On the other hand, the topic of housing is associated with diverse sets of disciplines ranging from architecture and engineering to psychology, sociology etc. and thereby, is an interesting topic for designers, planners, developers and policy makers who attempt to address the community needs by providing better housing situations (Dekker et al., 2011).

The prominent aim of most housing studies is to examine residential situations in order to understand housing demands and future consequences on the surrounding urban environment. Most research on housing satisfaction has been performed in socio- behavioral sciences and has looked at housing satisfaction as an essential explanatory input to study migration patterns, affordable housing and residential mobility. Apart from large governmental housing studies, most other studies narrow their research to get more detailed findings. Among those, some have focused on specific socio- economic groups of residents such as low income families (e.g., Addo, 2016), senior citizens (e.g., Temelová & Dvořáková, 2012), ethnics groups (e.g., Hall & Greenman, 2013), migrant workers (Tao et al., 2015), etc. In addition, several studies have examined satisfaction in a certain type of houses, such as those who have focused on public housing (e.g., Mohit & Nazyddah, 2011), informal housing settlements (e.g. Mitchell et al., 2018) and low cost housing (e.g., Makinde, 2015). Nevertheless, in some studies, housing satisfaction has been assessed to understand residents’ subjective well- being (e.g. Fernández-Portero et al., 2017; Aragonés et al., 2017).

4.2.2 Theoretical concepts

Housing satisfaction is a cognitive, multi-dimensional concept, which involves a process of complicated judgments and can be assessed through various methods. The satisfaction process starts from a comparison between what we have and what we want to have and several key theories have been formulated based on this fact. An example is “gap theory” (Galster, 1987) which is centered around the gap between the actual and desired housing situation and the theory of “person–environment fit” which has been applied by Kahana et al. (2003) to explore residential satisfaction. Other studies use similar theories to explore the consequences of housing dissatisfaction on the patterns of migration and residential mobility (Jiang et al., 2017). In addition to these theories, the concept of adaptation and housing adjustments serves a different perspective and explains the process of matching residential needs with characteristics

Housing well-being of a house during a period of time. This can result in reducing the deficits and can improve residential satisfaction (Wang & Wang, 2016).

In general, housing satisfaction is theorized based on the degree of which actual housing performance meets the household current needs and aspirations (Ogu, 2002). It contains numerous factors with a complex relationship among them. However, the extent to which these factors contribute to satisfaction is not clear, and the interrelation among these factors needs to be explored based on the contextual differences. Therefore, prior to discussing methodological approaches, we should define the main contributors to housing satisfaction, which can be supported by the existing scholarly literature.

4.2.3 Factors determining housing satisfaction (Indicator sets in housing research)

Studies on housing satisfaction develop various indicator sets according to the purpose and rationales of that specific research. Amongst these indicators, individuals’ socio- demographic attributes, dwelling’s characteristics and perceptions of housing conditions have been the most widely used factors. To study these main attributes, we classified them based on their relevance and popularity in the scientific literature and study them separately. Therefore, four categories of attributes starting from socio-demographic factors and then length of residency, ownership and cost and finally dwellings’ physical attributes are discussed in more detail.

4.2.3.1 Socio-demographic factors

Different households have different expectations and desires about their house. These expectations come partly from household demographic characteristics such as age composition, household size, household income etc. According to the literature of housing, socio-demographic and economic characteristics of households play a considerable role in housing satisfaction. Age, gender, marital status, education, employment status, income, number of households and length of residency are the most studied indicators. However, there is no consensus among the findings of researchers in terms of the positive or negative effects of these factors on housing satisfaction. Among the personal and household socio-economic variables, income has proven to have the clearest effect on households’ satisfaction, since it has a direct relation with obtaining desired material life and as a result enabling residents to have

Housing well-being suitable housing. Therefore, higher income usually contributes to higher level of satisfaction (e.g. Huang et al., 2015).

Education level is among the controversial factors, Lu (1999) and Oh (2003) referred to education as an insignificant factor, while some researchers (e.g. Vera- Toscano & Ateca-Amestoy, 2008) mentioned that higher education level has a positive influence on housing satisfaction as it can contribute to a better employment and economic situation. In contrast, being well educated might also increase expectations of living standards and accordingly brings lower satisfaction. Age is another relevant variable in the housing studies, as different generations have different needs. Age differences also address broader differences in life course and physical decline that can contribute to certain type of needs. However, age of the households referred as both significant and insignificant predictor in different housing studies (e.g. Baum et al., 2010; Mohit et al., 2010). For instance, Salleh et al. (2011) advocated that older tenants have higher satisfaction with their housing conditions than younger tenants, which is also confirmed by and Dekker et al., (2011). However, there is another argument that mentions older people spend more time at their houses so they would expect to have higher expectations of their house, which can contribute to lower residential satisfaction. Regarding other age groups, results from several studies state that age differences among young and middle-aged have not significantly contribute to a higher or lower residency satisfaction.

Gender differences based on the assumptions regarding the biological aspects has been examined in many built environment studies, and particularly in relation to physical activity and perceived safety (e.g. Amole, 2010). Hence, most gender studies focused on spatial issues such as way finding, crowding, density etc. rather than preferences on housing and other physical constructs (Amole, 2010). Very few studies found significant differences between males and females in terms of housing preferences. For example, Foye (2017) found that there is a positive but weak relationship between house size and men’s well-being. Also Aiello et al. (2010) argued that males are less satisfied with their housing than females, which was later confirmed by the study of Ibem and Amole (2013). The higher level of housing satisfaction among females has been explained through higher coping abilities and the stronger emotional attachment they can make with their environment. Marital status has also been proven to be an influencing parameter, and married households have shown a higher level of satisfaction compared with single individuals who have never got married. Employed

Housing well-being people, even those with a part time job may show a higher level of housing satisfaction as having a job can provide a more stable financial situation for them in compared with those who are unemployed. Despite this fact, most previous studies have not found any significant relationship between employment status and housing satisfaction, and even some studies have observed a negative correlation among them. Family size and number of children can tell a lot about the socio-demographic position of households and can determine satisfaction, as many scientists believe that an increasing number of family members can lead to decreasing housing satisfaction (Theodori, 2001).

4.2.3.2 Length of residency

Length of residency is normally studied among socio-demographic factors, but it has received special attention in many studies, and been found as a factor with significant influence on housing satisfaction (e.g. Mohit et al., 2010). Living duration is associated with the development of sense of belonging to the residency, which is an important human need and has been listed in Maslow's hierarchy of needs (1954). This has been shown to be consistent with the theory of adaptation and previous research on housing among various groups such as adults, ethnic groups, migrants etc. (e.g. AhnAllen et al., 2006; den Besten, 2010). In brief, length of residency can develop a sense of attachment to the dwelling (Adriaanse, 2007), and enhance the coping abilities of the occupants. Therefore, it can be counted as an important indicator in housing satisfaction measurements. On the other hand, it is inevitably related to home ownership, as home owners generally stay longer than renters (Diaz-Serrano, 2009). In other words, length of residency can be a sign for stable financial situation, which can increase the likelihood of satisfaction. However, as many researchers have already mentioned, living longer can also decrease the housing satisfaction, because of the dropping quality of the dwelling by aging and also changing the needs of the occupiers over time (e.g. Wang & Wang, 2016).

4.2.3.3 Ownership status and housing cost

Home ownership has been found to be an influential factor in residency evaluation since it provides residents with sense of control and security (Jussila et al., 2015). Lu (2002) argued that house renters have less control over their house in terms of changes, repair and maintenance, and as a result housing satisfaction has a lower level among renters. On the other hand, renters have fewer ties to the house as they feel it is a temporary place, while owners can establish a strong sense of stability and higher satisfaction 45

Housing well-being during the years. Also Elsinga and Hokstra (2005), who did a research on housing satisfaction in eight European countries, found that owner-occupiers are happier than tenants in general, and just in one of those countries, satisfaction was not significantly different among owners and tenants.

Many authors tend to support a positive relation between cost of housing and satisfaction, as higher cost houses can have a better quality (e.g. Ismail et al., 2018). At the same time, living in a low cost rental houses should not necessary bring lower levels of satisfaction as Ibem and Aduwo (2013) reported that residents in low cost public houses were mostly satisfied with their houses’ conditions. However, cost of housing and rent price of a house can be a reason for dissatisfaction as well. Accordingly, cost should be matched with the condition of the dwelling and if a tenant feels that there is a possibility of finding a better quality house with the same price, or even same quality and lower price, then this would lead to dissatisfaction. Additionally, cost of housing could cover monthly maintenance costs, which can be relevant for apartment houses and those with shared spaces. These kinds of costs also can be seen as unfavorable costs at some cases and can decrease housing satisfaction.

4.2.3.4 Dwelling physical characteristics

Physical attributes of a dwelling are major contributors to housing satisfaction and poor physical house characteristics can result in certain level of dissatisfaction. These attributes include structural attributes such as type, age, size, design, exterior space, interior quality, number of bedrooms, safety/security facilities and parking space of house (e.g. Baum et al., 2010; Hipp, 2010). Among those, size, type and design of the house are reported to have profound influence on the housing overall satisfaction. For instance, Chen et al. (2013) found that bigger size and better housing facilities can influence housing satisfaction positively. Roberts-Hughes (2011) concluded that small sizes house and shortage of space have resulted in dissatisfaction and consideration for moving home for some residents. Jiboye (2014) found that type of house has a significant impact on residential assessments. Aulia and Ismail (2013) also found that type of residence is an important physical factor among middle-income households. Buys and Miller (2012) concluded that satisfaction with dwelling design play a significant role in residential satisfaction particularly for higher density housing. Table 4.1 shows the major contributors to the house satisfaction in the selected recent literature.

Housing well-being

4.2.3.5 Subjective evaluation of all the house objective characteristics

The vast majority of housing studies have focused on the physical characteristics (functionality and aesthetics) of dwellings including housing cost and ownership status, while some other have addressed the perceived satisfaction of households. Recently, some empirical studies have used both objective characteristics and subjective evaluation of residents to assess the degree of house satisfaction (e.g. Sealy–Jefferson et al., 2017). This approach reflects the effects of dwelling conditions on the satisfaction, and thus can provide more information on what residents like or dislike about their house.

Table 4.1 Different literature regarding the major contributors to the house satisfaction

Housing Variables Socio-demographic attributes Dwelling characteristics attributes

Scientific literature

(Author’ s name)

status

berbedrooms of

Age Gender Education Maritalstatus Householdsize Householdincome Lengthresidency of Ownership Typeof House Sizeof house InteriorQuality ExteriorSpace Num of Price housing Design Ageof house Safety/securityfacility Parking

Fernandez-Portero * * * * * * * * * * * * * et al. (2017) Azimi & Esmaeilzadeh * * * * * * * * * * * * * (2017) Wang & Wang * * * * * * * * * * * * (2016) Huang et al. (2015) * * * * * * * * * * * * * Pekkonen & Haverinen- * * * * * * * * Shaughnessy (2015) Zumbro (2014) * * * * * * * * * * * * * * * * * Mohit & * * * * * * * * * * * * * * * Nazyddah(2011) Li &Sung (2009) * * * * * * * * * * * * * * * * Ioannides & * * * * * * * * * * * * * Zabel(2008) Vera-Toscano & Ateca- * * * * * * * * * Amestoy (2008) * *

Housing well-being

4.3 Multiple regression analysis

In order to explore the implicit effect of various house dimensions on the residents’ overall assessment of their house multiple regression analysis was conducted with overall house satisfaction as the dependent variable and 31 independent variables (including socio-demographic variables). The reason that we first used this method is that regression serves as a primary method to shed light on more fundamental relations without addressing the complex relationships among variables. Also, the reason that we chose regression from various alternative methods (ordered logit, etc.) is that the estimated responses were checked and were positive.

We assumed that the scales, which used to measure satisfaction have interval properties (same degree of difference between items) based on many studies that utilized such scales (e.g. Diaz-Serrano & Stoyanova, 2010). Moreover, as the number of respondents is relatively low compared to the number of variables, we have to reduce the number of estimated variables. So, all multiple categories were merged into binary categories. House size was divided into “Small house” and “Non-small house”. House type was divided into “Flat” and “Non-flat”. Age of house includes “New house” and “Old house”. “Exterior space” was divided into “With exterior space” and “No exterior space”. Monthly price of housing was merged into “Low cost house” and “High-cost house”. House view was divided into “Green view” and “Non green view”. Design/layout refers to “Design fits” and “Design does not fit lifestyle”. Safety facility divided into “With security facility” and “No security facility”. Renovation situation contains “Needs repair” and “No need to repair” and parking situation was divided into “Free parking” and “No free parking” (Table 3.6).

Then, to examine our conceptual model by multiple regression analysis, the overall house satisfaction treated as the dependent variable and examined by 31 independent variables. Since one of our regression assumptions is that the residuals (prediction errors) are normally distributed, therefore, checking the histogram of residuals allows us to check the extent to which the residuals are normally distributed. The well-behaved residuals can lead to correct P values and thus are very important for any linear regression analysis. Our histogram (Figure 4.1) shows a fairly normal distribution. Thus, based on these results, the normality of residuals assumption is met.

Furthermore, the normal P-P Plot of regression residuals (Figure 4.2) can also be used to assess the assumption of normally distributed residuals. If data points (little circles) follow the diagonal line, the residuals are normally distributed. Our P-P plot shows 48

Housing well-being

Figure 4.1 The Histogram of the overall housing satisfaction

Figure 4.2 Normal Probability Plot (P-P) of the regression standardized residual dependent variable for overall house satisfaction

Housing well-being minor deviation and thus the normality of residuals assumption is satisfied. The results about R-squared (R2) and adjusted R-squared are shown in Table 4.2. It shows that the indicators accounted for 80 percent of variance of the overall house satisfaction.

4.3.1 Interpretation

Regression coefficients in this analysis represent the magnitude of the influence of the explanatory variables on the residents’ overall house satisfaction (Table 4.3). Among all of the indicators in this analysis, ten variables show a significant relationship with the overall house satisfaction (P<.05). From the socio-demographic variables just “Highly educated” variable have a strong positive correlation with the dependent variable. “New house” variable from the objective characteristics of the house also is significantly related to the “Overall house satisfaction”. The other significant predictors are all from the subjective category and included “Ownership satisfaction”, “Design satisfaction”, “House-type satisfaction”, “Safety satisfaction”, “Exterior space satisfaction”, “Renovation situation satisfaction”, “Parking satisfaction” and “House size satisfaction” (Table 4.3).

Among the significant predictors, the only variable that shows a negative relation with the overall house satisfaction is “New House” variable (-.10), which indicates the houses that were built after year 2000. Unexpectedly, “Highly educated” variable shows a significant relationship with overall house satisfaction (.08). The significancy is mainly due to the fact that this variable presents a large group of samples (residents with any collage degree), which is almost 81% of our entire sample size. Apart from this variable, the other independent variables that have strong positive correlation with the overall housing satisfaction are “House-type satisfaction” (26), “Design satisfaction” (.20) and “Ownership satisfaction” (.19), respectively. In contrast, “Renovation situation satisfaction” influences the overall house satisfaction with the lowest effect size (.10), following by “Parking satisfaction” (.09). The results clearly indicate the importance of subjective variables compared with the objective components.

Although multiple regression analysis provides some insights into the impact of socio- demographic characteristics and all the mentioned house-related characteristics on overall house satisfaction, but this method is unable to capture the indirect relationships and the complex associations among these variables. Thereby, further analysis is

Housing well-being

Table 4.2 Goodness-of-fit of the regression model of overall house satisfaction

R R2 Adjusted R2 Std. Error of the Estimate

.895 .801 .766 .632

Table 4.3 The estimated parameters of the house satisfaction using multiple regression analysis

Unstandardized Standardized

Coefficients Coefficients

Model Parameters B Std. Error Beta t Sig.

(Constant) .107 .513 .207 .836

Age -.001 .003 -.017 -.402 .688

Socio- Household income .045 .032 .058 1.382 .169

demographic Male -.047 .097 -.018 -.477 .634 indicators Single person -.021 .129 -.008 -.163 .871

Highly Educated .270 .124 .082 2.173 .031

Household Size .045 .057 .041 .784 .434

Length of residency .049 .041 .046 1.198 .233

Number of bedrooms -.015 .051 -.010 -.287 .774

Flat .260 .156 .098 1.665 .098

Small house -.158 .137 -.060 -1.147 .253

Objective New house -.262 .127 -.100 -2.056 .041

characteristics No exterior space .006 .167 .002 .036 .971

Low cost house .139 .107 .053 1.297 .196

Needs repair .069 .122 .022 .569 .570

Green view house .063 .121 .022 .518 .605

No security facility .062 .101 .023 .614 .540

Design does not fit lifestyle -.127 .148 -.032 -.861 .390

Rental house -.169 .150 -.064 -1.127 .261

Free parking -.157 .140 -.060 -1.121 .264

Parking satisfaction .037 .014 .107 2.641 .009

Ownership satisfaction .161 .049 .185 3.250 .001 51

Housing well-being

House-type satisfaction .227 .051 .262 4.468 .000

Subjective House size satisfaction .118 .053 .141 2.241 .026

evaluation Number of bedrooms -.017 .039 -.024 -.440 .660 satisfaction

Exterior space .082 .028 .141 2.897 .004 satisfaction

Age of house satisfaction -.003 .038 -.003 -.070 .944

House view satisfaction -.013 .036 -.017 -.364 .716

Renovation situation .076 .038 .091 2.014 .045 satisfaction

Monthly price satisfaction -.021 .032 -.029 -.646 .519

Safety satisfaction .138 .041 .139 3.339 .001

Design satisfaction .178 .044 .208 4.033 .000

necessary to understand the mechanism of effects among the study variables, which will be explained in the next part.

4.4 Analysis and results: Path analysis

The key objectives of this chapter are to evaluate and understand the interrelationships among the study variables in formulating house satisfaction. A path analysis is deemed to be an adequate solution for the achievement of these objectives, thus it is used in this chapter. By this method, it is possible to conduct simultaneous assessments of the associations among multiple dependent and independent attributes, which can increase the statistical efficiency (Hair Jr et al., 2016).

4.4.1 Conceptualization

Based on the aforementioned insights from different housing studies and well-being research, and the relevance of these studies to investigate the link between different house attributes (objective and subjective) and socio-demographic variables as well as the overall subjective assessment of house, a conceptual model is elaborated and will be discussed in this section. To develop the conceptual model, first the main determinants of housing satisfaction categorized in three different groups (Figure 4-3). The first category is the socio-demographic variables of the residents including the

Housing well-being

Socio demography variables Objective characteristics

Subjective evaluation

Overall satisfaction

Figure 4.3 The main categories of determinants of housing satisfaction

length of residency. The second category is the objective characteristics of a house including ownership, cost of housing and all the physical components of the dwelling. And finally the third category is the satisfaction with each of the objective characteristics. All of these variables together influence overall housing satisfaction within our conceptual framework. So overall housing satisfaction will be studied by exploring the relationships between the degree of housing satisfaction with a set of personal/household characteristics, objective characteristics of the house and subjective evaluation of these characteristics. Here, personal/household characteristics are associated with different needs of people, while house characteristics indicate the extent by which a house offers the opportunity to satisfy these needs. Figure 4-4 presents the conceptual model of this study in more details.

4.4.2 Path analysis for housing satisfaction

Our conceptual model was estimated using AMOS 22 software package with the Maximum Likelihood Estimation (MLE) method. This method assumes multivariate normality, which is a robust method. It enables estimating the direct and indirect correlations among factors, the p-values of the paths including the standardized regression weights of the conceptual model. In the first step, all of the attributes were included in the model to see whether they have any direct or indirect relation with the exogenous variables. After omitting the insignificant relationships from the path model,

Housing well-being

Figure 4.4 The conceptual model

age, length of residency and household income from the socio-demographic group were remained. However, Age of respondents and length of residency were highly correlated which causes less accuracy for path coefficient. Thereby, due to the multicollinearity, one of them should be deleted. Since length of residency variable is theoretically more meaningful for housing satisfaction, it remained and age variable deleted from themodel. Another strong correlation among the significant variables in the same group also found among “House size satisfaction” and “Type of house satisfaction”. It is also theoretically meaningful, as small houses are usually flat houses and large houses are

Housing well-being

Figure 4.5 Results of the path analysis for house satisfaction

villa. Therefore, again due to the multicollinearity one of the variables should be omitted. Considering the theoretical and statistical relevance of “Type of house satisfaction”, it stayed and accordingly “House size satisfaction” omitted from the model (Figure 4.5).

The results of the final path model (Table 4.4) demonstrate that the model excellently fit to the data. Chi Square divided by the model degrees of freedom (/df or CMIN/DF) is an essential measure, which should to be less than 2 and preferably close to 1 (Ullman, 2006). This indicator is equal to 1.43 for our model. In addition to that, the Root Mean Square Error of Approximation (RMSEA) should be less than .08 and ideally less than .05 to represent a great fit (Washington et al., 2010) and in our model RMSEA value is .045. Other indicators of a good fit as mentioned by many researchers (e.g. Moss, 2009) are indices that compare target model with null model, such as Comparative fit index (CFI), Incremental fit index (IFI), Tucker Lewis index (TLI), Normed fit index (NFI) and Goodness of Fit Index (GFI). If these indexes exceed the amount of .90 the model tends to generate an acceptable fit. Checking all of these indicators shows that our model perfectly fits. Moreover, R squared (R2) of the ultimate endogenous variable (called squared multiple correlation value in AMOS) is another important criteria signaling a good model. 55

Housing well-being

Table 4.4 Goodness of the fit of the model

Fit index Estimate

Chi-square (CMIN) 1.43

Degrees of freedom (DF) 28

Chi-square / degrees of freedom 40

Root mean square error of approximation (RMSEA) .045

CFI .98

IFI .98

TLI .97

NFI .95

Our proposed theoretical model satisfactorily accounts for the 73 percent of the variance in residents’ overall house satisfaction (R2=.73). Statistically significant relationships are found among the main model variables as below (refer to Table 4.5):

 “Type of house satisfaction” has the largest direct effect on the overall house satisfaction (.32).

 “Design satisfaction” has the second direct and strong correlation with the effect size of .28, followed by “Ownership satisfaction” (.21), “Exterior satisfaction” (.20) and “Safety satisfaction” (.15).

 Among the indirect relationships, “Household income” has the strongest influence on the overall house satisfaction (.17).

The other main results can be summarized as:

 Overall house satisfaction is negatively and indirectly influenced by “Rental house” (-.09), “Low cost house” (-.05) and “No security facility” (-.04).

 In the objective characteristics group, “Low cost house” significantly influences “Design satisfaction” with negative effect (-.19). As expected, “Rental house” show significant negative effect on “Ownership satisfaction” (-.24). Likewise the correlation between “Rental house” and “Exterior space satisfaction” is negative (-.22). Moreover, “No security facility” strongly impacts “Safety satisfaction” (-.30), which indicates the importance of safety facilities for feeling safe at home.

Housing well-being

 From the socio-demographic variables “Household income” and “Length of residency” are indirectly related to the overall housing satisfaction (with the effect size of .02 and .17, respectively). “Household income” also shows significant effects on “House-type satisfaction” (.22) and “Ownership satisfaction” (.20). “Length of residency” only shows strong correlation with “Rental house” (-.27).

4.4.3 Interpretation and discussion

Generally speaking, our results support the main parts of the initial conceptual model, and are also coherent with the previous studies. Although many of the variables from the initial model have been insignificant and omitted, but all of the key aspects have been remained and showed significant influence on the overall housing satisfaction. Below are the main findings and interpretations on the results:

 House-type satisfaction, design satisfaction and ownership satisfaction have the strongest influence on the overall evaluation of house, which is in line with the findings of other researchers (e.g. Buys & Miller, 2012; Jiboye, 2014). Being satisfied with house-type can predominantly reflect a large portion of housing satisfaction. It is due to the fact that house-type is more than just a dwelling form. It is linked with lifestyle and socio-cultural factors that reflects the ideals of occupiers including their expectation regarding certain house qualities. House-type satisfaction is directly associated with household income (.22), since certain income-level households look for certain type of house and to some extent house type can reflect the income level of households. The degree of satisfaction with dwelling design can also highly reflect the extent to which house is adequate for performing different activities and can fulfill the occupants’ needs. House design can basically tell whether different space features of house are satisfying and thus implying satisfaction with house size, number of bedrooms satisfaction, style of the house (age of house) etc. Satisfaction with home ownership has been found to be an influential factor in residency evaluation. It is highly influenced by household income (directly and indirectly), as certain amount of income leads to property ownership. Negative impact of “Rental house” on “Ownership satisfaction” again emphasizes on the renters’ dissatisfaction with their ownership status. Moreover, our results indicate that “Length of residency” indirectly impacts “ownership satisfaction”

Housing well-being

Table 4.5 Standardized path estimates of direct, indirect and total effects

Dependent Paths Direct Indirect Total variables effect effect effects

Length of residency  OHS .00 .25 .25

Household income  OHS .00 .17 .17

Overall Rental house OHS .00 - .09 - .09

house No security facility  OHS .00 -.05 -.05 satisfaction Low-cost house  OHS .00 -.05 -.05 (OHS) HTS  OHS .32 .00 .32

DS  OHS .28 .00 .28

OS  OHS .21 .00 .21

ESS  OHS .20 .00 .20

SS  OHS .15 .00 .15

Ownership Household income  OS .20 .09 .29

Satisfaction (OS) Rental house  OS -.24 .00 -.24

Length of residency  OS .00 .06 .06

Design Low-cost house  DS -.19 .00 -.19

Satisfaction (DS) Household income  DS .00 .06 .06

House-type Household income  HTS .20 .00 .20 Satisfaction

(HTS)

Safety Household income  SS .00 .06 .06

satisfaction (SS) No security facility  SS -.30 .00 -.30

Exterior space Rental house  ESS -.22 .00 -.22 satisfaction Household income  ESS .00 .08 .08

(ESS) Length of residency  ESS .00 .06 .06

Rental house Household income  Rental house -.36 .00 -.36 Length of residency  Rental house -.27 .00 -.27

Low-cost house Household income  Low-cost house -.33 .00 -.33

No security facility Household income  No security facility -.20 .00 -.20

Housing well-being

(.06) which is obvious, as long-term residency is usually bond with house ownership and rental contract are usually temporary.

 According to our result tenants have lower house satisfaction. Other researchers (e.g. Elsinga & Hoekstra, 2005; Jussila et al., 2015) also found the similar results, as renters have less control over their house in terms of changes, repair and maintenance and also have fewer ties to the house. Also, living in rental house is negatively correlated with exterior space satisfaction, which might be due to the fact that the majority of rental houses are apartments with little or no exterior space.

 The negative influence of “Low cost house” on “Design satisfaction” (-.19) indicates that these houses just meets the basic needs of dwellers and provide

less design-related qualities. Actually, living in a low cost house can negatively influence the overall house satisfaction in an indirect relationship (.05).

 The indirect effect of household income on housing satisfaction has clearly been proven, since it has a direct relation with obtaining desired material life and as a result enabling residents to have a suitable housing.

 “Length of residency” positively influences the overall assessment of the house, which might be due to the sense of attachment and coping ability of households.

4.5 Conclusions and discussion

This chapter has aimed to enhance the understanding of the housing satisfaction concept and how this major life domain is affected by socio-demographic characteristics of individuals as well as the characteristics of one’s house. Path analysis was used to capture the links between various variables and their impacts on overall housing satisfaction. The findings highlight the major influence of house-type satisfaction, design satisfaction and ownership satisfaction on feeling satisfied with the house. Owner-occupiers are happier than tenants in general which stem from the fact that people have a natural preference for owning their territory apart from other economic and social reasons behind the desire for ownership.

Type of house also reflects many qualities of the house including the dependency and freedom for activities. The importance of house design in compared with house size has been particularly reflected in small lofts of “StrijpS”, one of our

Housing well-being study area in Eindhoven (See Chapter 3), where innovative and smart design enables residents combining working and living in multifunctional spaces. Moreover, this study shows that living in affordable houses does not matter as long as the house design satisfies the occupant’s needs. Although household income shows a significant contribution to house satisfaction and influences almost all aspects of housing, the dominant impact of income is on ownership satisfaction. This partly reflects the fact that people usually buy a property based on the type and design of the house. Overall, the residential satisfaction is complex in its nature. It is bonded to economic, social and physical structure of the community and at a higher level to the housing policies. Furthermore, housing satisfaction is connected to its surrounding and getting influence from the area it is located in, thereby our next chapter will address neighborhood satisfaction.

Neighborhood well-being

5.1 Introduction

Neighborhood is the second immediate space after housing that urban residents’ encounter during their life course. It is the most basic spatial unit that provides opportunities for the fulfillment of its residents’ daily needs and social interaction (Ettema & Schekkerman, 2016). A vibrant neighborhood can shape the activity patterns of residents and make their life more efficient, pleasant and relaxed. It can also be considered as the most integral part of its residents’ life by offering different infrastructures and mixture of facilities, which furthermore will act as a source for physical, recreational and social activities. From a physical point of view, neighborhoods provide access to all amenities, high quality roads, pedestrian and bike paths to generate a connected and functional space and promote an active lifestyle. Neighborhood environmental condition such as good mixture of trees and greenery

Neighborhood well-being including clean streets, water and air with no traffic congestion and noise pollution influence residents’ physical and mental health positively. In addition, from a social perspective, by bringing residents together in local public spaces, providing a safe area for all residents to participate in activities and having contact with neighbors, neighborhoods can produce a sense of attachment and be a meaningful source for identity. Thereby, it is said that satisfaction with various neighborhood features associate with overall satisfaction of neighborhood, which can play a significant role in the assessment of quality of life (e.g. Węziak-Białowolska, 2016). In other words, the central position of the neighborhood in providing and facilitating life with different elements explains why neighborhood satisfaction can be a significant predictor of life satisfaction (Marans, 2015).

Because of this important role, many researchers have explored the associations between various neighborhood characteristics, residents’ satisfaction and well-being, (e.g. Zhang & Zhang, 2017 ). However, due to the complexity of neighborhood research and its broad nature, the vast majority of studies have focused on a particular aspect of neighborhood rather than the whole diverse set of characteristics such as the studies on neighborhood safety and social attachment, which in turn created a substantial body of research in environmental psychology and social science. Nevertheless, not all the studies of the neighborhood satisfaction revealed aligned results, which can be due to the contextual differences. The existing contradictions in the literature brought uncertainty in choosing the relevant attributes for future studies. In addition, researchers believe that even similar physical circumstances can be perceived differently across individuals (Hipp, 2016). However, there is still a lot of room for systematic investigation of this multifaceted concept by considering all domains and subdomains.

In formulating the conceptual framework of this thesis (see chapter 2), neighborhood satisfaction contributes to quality of life as a main domain, and thus, this chapter will focus on neighborhood satisfaction. Explicitly, the following research questions will be addressed:

I. What are the main predictors of neighborhood satisfaction? Or which neighborhood features have received empirical support in predicting neighborhood satisfaction?

II. To what extent do these indicators contribute to predicting neighborhood satisfaction?

Neighborhood well-being

III. What are the relationships among these indicators and to what extent do they affect each other?

5.2 Literature review

Neighborhood is a multidimensional phenomenon, which is beyond a static fixed geographical unit, but rather a dynamic concept, comprising a network of associations among people and places (Gardner, 2011). Many researchers attempted to define neighborhood in different fields of study in order to provide a basis for their research. As a result, the definition evolved over time and multiple terms and definitions developed during the past several decades.

5.2.1 A place called neighborhood (terminology and definitions)

The classic definition by Lewis Mumford (1954) considered neighborhood as a natural urban unit: “when many functions of cities tend to be distributed neighborhoods formed naturally”. In its classic definition neighborhood is geographically a limited area composed around a center with clear edges and a balanced mix of activities. Later, the emergence of other concepts such as sliding edged neighborhoods and multi-center neighborhoods transformed this classical definition. Through approaches such as cognitive mapping, researchers realized that people of the same area have different perception regarding their neighborhood size and its boundaries, which is dependent on their residential location as well as their own profile (life course and socio-demographic characteristics). Following this result, Lee (1968) defined a concept called “socio-spatial schema” where important physical elements of neighborhood together with social relationships can depict a different schema for each resident, and neighborhood is the combination of all residents schema. Moreover, Chaskin (1995) identifies three dimensions for neighborhood: neighborhoods as social units, neighborhoods as spatial units, and neighborhoods as a network of associations.

As many authors argued (e.g. Foster & Hipp, 2010; Koohsari et al., 2015), there is a mismatch between the administrative definition of a neighborhood and residents experience of a neighborhood, which also are known as: territorial neighborhood and ego-centered neighborhood, respectively. Territorial neighborhood refers to administrative fixed boundaries while ego-centered neighborhood reflects resident- oriented area that is centered on individual’s unique mobility pattern, exposures and life style. This concept can be the extension for “activity space” concept (individual's day-to-

Neighborhood well-being day lived experience) by Albert and Gesler (2003), where they theorized neighborhood as “the local areas within which people move or travel in the course of their daily activities”. Among the many definitions, it seems that Meegan and Mitchell (2001)’s definition includes all physical, social and psychological concepts on neighborhood as: “Neighborhood is a key living space through which people get access to material and social resources, across which they pass to reach other opportunities and which symbolizes aspects of the identity of those living there to themselves and to outsiders.” Given this definition, neighborhood satisfaction can be a process consisting of individual definitions of neighborhood, their personal characteristics and their very subjective evaluation.

As terminology involves with concepts and definitions, consequently several terms have been used to explain this concept, among those; community, district, area and vicinity are the most common terms that used interchangeably. The challenge of setting boundaries for neighborhood is also reflected in various terminologies and terms which each of them can emphasize on different geographical or social aspects and dimensions to identify what is called as neighborhood. For instance, the geographical and physical boundaries such as highways, main roads, rivers etc. can define a district, whilst the area that surrounds the house with local, familiar residents (usually in a same social class) is called community. Even by using “neighborhood” people can mean different things, Marans and Rodgers (1975) extended the definitions and suggested “micro neighborhood” and “macro neighborhood” terms, where the former is the adjacent area to the house contains around 8 blocks and the latter is the official districts with all the facilities. In Dutch terminology micro neighborhood that covers daily activities of residents is called as “buurt” while a larger area with more facilities is called as “wijk” (Wassenburg, 2013).

5.2.2 Research background

Community research has a long history in planning, social science and related fields. Precisely, the emergence of neighborhood level studies has begun a century ago where researchers became interested in investigating the effects of place on human and collective living. Over the past few decades, many researchers attempted to explore the neighborhood characteristics and the mechanism of its contribution to residents’ well- being (e.g. Amérigo & Aragones, 1997; Talen, 2000; Austin et al, 2002; Chapman &

Neighborhood well-being

Lombard, 2006; Santos et al., 2007; Sirgy et al., 2013; Cao, 2015; Hadavi, & Kaplan, 2016; Zhang & Zhang, 2017).

Substantial effort has been devoted to explore the associations of health (e.g. Rollings et al, 2017), crime (e.g. Cozens & Love, 2015), personal attachment and network (e.g. Dassopoulos & Monnat, 2011; Clark et al., 2017) with neighborhood satisfaction. The fundamental shift from physical to social characteristics of neighborhood resulted in employing methods to assess satisfaction of both physical services and social conditions. Consequently, different concepts such as social cohesion and sense of community emerged and created a new line of research with a focus on evaluation regarding crime, stigma and neighborhood image.

In fact, there are two separates lines of research that recently overlapped. One line, which was developed by urban scientists, is related to planning and design principles and service-oriented communities. The other line mainly focuses on social structure aspects such as cohesion, stigma and sense of community. The overlap of these two research arrays has been reflected in interdisciplinary fields and concepts such as “New urbanism” in which both dimensions intertwined. New urbanism shows the ability of the built environment in creating sense of community by identifying some principal components such as permeability, accessibility, walkability and social interaction (Talen, 2008). However, there are different criticisms on its ability to comprehensively capture the social layers of neighborhoods and therefore investigating other theoretical concepts is inevitable.

5.2.3 Theoretical concepts

There is no single theory that can explain neighborhood satisfaction, mainly because both person and environment are fluid and affect each other simultaneously, which makes it difficult to identify the cause and the effect (Félonneau & Causse, 2017). But, a series of theories have dealt with human contentment with environment. Among those, theory of place, affordance theory and theory of human motivation have provided the foundation for numerous evaluation studies (e.g. Raymond et al., 2017).

Theory of place in environmental psychology is pioneering in defining the experience of pleasure deriving from place and contains three main psychological stages: cognition, affection and behaviors (Rosenberg & Hovland, 1960). Affordance theory by psychologist James Jerome Gibson (1979) emphasized human evaluation of environment through possibilities of operation and action. This theory can have many

Neighborhood well-being implications in different fields and some authors expanded it to the built environment with respect to functionality in person–environment relations (e.g. Raymond et al., 2017). The theory of human motivation or Maslow’s hierarchy of needs has adopted by a variety of disciplines. It can also be well adapted to residents’ needs at the neighborhood setting, as the general human needs can be extended to individuals and collective needs in a community. Thereby, different levels of residential needs can be organized hierarchically from basic needs to higher-order needs. The consistent concept across all of these theories is that judgment and evaluation are more salient than reality or objective components of the environment as satisfaction is a psychological construct.

5.3 Factors determining neighbourhood satisfaction

Based on the review of existing literature and available theories, numerous factors can influence individuals’ evaluation of neighborhoods. However, due to the complexity of the nature of the concept and the context-dependency, as expected, there is mixed evidence on the relevance of some of attributes such as neighborhood facilities, density, housing type, etc. (Lovejoy et al., 2010). Nevertheless, based on the empirical evidence some variables such as safety (e.g. Lee et al., 2017), social attachment (e.g., Di Masso et al., 2017), upkeep (Hur & Nasar, 2014) and age (Zhang & Lu, 2016) are consistently associated with the neighborhood satisfaction. It is also good to note that neighborhood is not a closed system but is linked to wider aspects of the regional and national system (Pacione, 2011). Therefore, the roots of most of the neighborhood problems must be sought beyond the neighborhood scale. Due to the limited scope of this research, we explore only the variables that are directly related to the neighborhood level evaluation.

5.3.1 Individuals’ socio-demographic characteristics

Age is one of the personal characteristics that is always among the influential variables in the neighborhood satisfaction studies. Many researchers (e.g. Zhang & Lu, 2016) found that older people have a higher level of neighborhood satisfaction compared to younger residents, which might be due to the fact that higher purchasing power enables them to move to their favorite neighborhood (Lovejoy et al., 2010). Being single can also impact neighborhood satisfaction but negatively, as it might affect feelings of safety and social interactions. As such, Galster and Hesser (1981) reported that single women were less satisfied with their neighborhood than others. Regarding gender, many studies found a weak impact on neighborhood satisfaction (e.g. Kweon et al., 2010). 66

Neighborhood well-being

Regarding other Socio-economic variables such as household income, education level, household size and presence of children, the results are very mixed and not consistent. For instance, Dekker and Bolt (2005) found that presence of children is positively associated with neighborhood interactions, which can consequently enhance neighborhood satisfaction, while Hipp (2009) reported that divorced households with children have lower neighborhood satisfaction than others. As a result, those socio- demographic variables that contribute to integrating more in neighborhood activities are likely to increase the social contacts in the neighborhood and tend to be influential on neighborhood satisfaction.

5.3.2 Social environment characteristics

A vast amount of literature emphasizes on the social environment as a key domain in judgments regarding neighborhoods and much empirical research has used social determinates to measure neighborhood satisfaction. However, since these variables should be matched with local context, different authors have chosen various sets of variables. Nevertheless, in the international context, there are some widely recognized variables such as safety and personal attachment that were showed to have strong connection with neighborhood satisfaction. Also, based on Maslow’s hierarchy of needs, safety and personal attachment are the second and third levels of needs. In the Dutch context, livability and social status have particularly been listed as relevant influential variables of neighborhood satisfaction (e.g. Van Assche et al., 2018). Hereafter, some of the most important social environment variables are discussed in more details.

5.3.2.1 Safety

Safety evaluation or fear of the crime is the most common used criterion in neighborhood satisfaction research and highly contributes to residents’ subjective assessments of neighborhoods. Also, safety has been linked to concepts such as “control over the environment”, and thereby neighborhood physical structure and street layout have received a lot of attention, due their impact on the feelings of safety. Accordingly, environmental criminologists investigated the impact of spatial factors on fear of crime. Significant research by David Lynch (1960) and Jane Jacobs (1961), highlighted the impact of urban design on safety and concepts such as “Defensible space” (Newman, 1972) and “Indefensible space” (Cozens et al., 2002) provided the intellectual foundation for in-depth crime analyses. It has been proven that the level of safety is different across different socio-demographic groups. For instance, women feel 67

Neighborhood well-being less safe than men and fear of crime is higher among older people. Also, feelings of safety are connected with stigmatized neighborhoods (Cozens, 2011) and can impact performing physical activities such as walking, which will be discussed later in this chapter.

5.3.2.2 Personal attachment

Personal attachment is another widely used attribute in neighborhood satisfaction studies. It is a two-dimensional factor comprising of place attachment and social attachment (Raymond et al, 2017). It can also be considered as a mixture of feelings that reflects the degree of belongingness, social trust and evaluation of cohesion (Williams & Vaske, 2003; Dassopoulos & Monnat, 2011). Principally, interaction with the local network or social ties bind people to specific places and can maintain higher level of personal attachment. Moreover, it has been found that age and personal attachment are significantly related as older residents usually have a longer residency and therefore, the familiarity with the area and people are higher among them (Cramm, & Nieboer, 2015).

5.3.2.3 Liveliness

Liveliness has also been listed among the variables related to social satisfaction of a neighborhood. It is a subjective indicator with links to some physical characteristics of area such as density, number of shops, position in city, and socio-demographic characteristics of the area. A livable area is full of life, excitement and energy and can offer opportunities for interaction with other residents, thereby can positively impact neighborhood assessment (Lovejoy et al., 2010). Different research studies have shown that people like quiet micro neighborhoods but prefer lively macro neighborhoods (a broader census-tract context) with lots of activities and events. This also has highlighted the connectivity factor and easy access to the city center and other major activity centers.

5.3.2.4 Image/reputation

Neighborhood image or its reputation demonstrates how residents and outsiders evaluate the socio-economic and cultural situation of the neighborhoods and differentiate them from surrounding areas (e.g. Müller, 2017). Generally, neighborhoods can be an identity for their residents, as Galster (2002) argued that socio-economic status of a neighborhood can provide “prestige” for its residents. Conversely, a negative 68

Neighborhood well-being image or poor neighborhood reputation can lead to a low level of neighborhood satisfaction and relocation decisions (e.g. Goetze, 2017). In the Netherlands, the debate about disadvantaged neighborhoods is often connected to unfavorable social status and stigma. Thus, neighborhood reputation should be considered an influential attribute in neighborhood satisfaction studies.

5.3.3 Physical and environmental characteristics

The physical characteristics of neighborhoods that potentially influence satisfaction are numerous. Employing the human needs hierarchy can provide the basis for what is considered by residents as the main needs in the neighborhood. These needs progress from the most basic needs such as access to clean air to higher-order needs that include accessibility, quality of green space and layout.

Following the essential physiological needs, access to clean air, water and land including food supply are the most important indicators for evaluating neighborhoods. They are followed by lower traffic volumes and good quality roads and paths that provide access to adequate services and facilities to fulfill people’s needs (Cao, 2016). Moreover, level of noise pollution and presence and quality of green landscape (Zhang & Zhang, 2017) are directly connected to physical and mental health of residents and are posited at the second level of the needs hierarchy. Architecture of housing and well- structured layout of buildings are physical qualities, which belong to a higher level of the needs hierarchy (Sirgy & Cornwell, 2002; Hur & Jones, 2008). It is said that if a neighborhood can fulfill the main needs of residents, then higher level of needs might be considered by residents (e.g. Alfonzo, 2005; Buckley et al, 2017).

5.3.4 Physical activity satisfaction

As physical activity is directly linked to health, the neighborhood characteristics that can support an active lifestyle are increasingly become important. In addition, several studies have emphasized the associations of environmental attributes with physical activity behaviors (e.g. Orstad et al., 2017). As a result, increasing attention is being paid to pedestrian-friendly urban forms and walkability (e.g. Buckley et al, 2017).

The Netherlands is well known for having compact cities and high density residential areas. These spatial characteristics including the culture of biking made most neighborhoods bike friendly. In contrast, countries such as the US have totally car- dependent neighborhoods, and in some cases without even pedestrian paths. Thus,

Neighborhood well-being physical activity related indicators are very context dependent. However, as in the Netherlands (our study context) walking and cycling are part of the daily activities, we should investigate them separately. In general, quality of path, traffic safety (risk of accident), safety regarding crime and greenery are the most relevant indicators to evaluate neighborhood physical activity support (Talen & Shah, 2007; Schepers et al., 2015; Jensen et al., 2017).

5.3.5 Satisfaction with services in neighborhood

Neighborhoods should respond to the main needs of their residents such as the need for access to services and amenities for grocery shopping, education, healthcare, recreational activities etc. Among those, food related facilities are the most important and the most frequent visited places in neighborhoods. Availability of amenities is especially crucial for older residents with physical and economic limitations (Rioux & Werner, 2011). Generally, residential proximity to facilities, quality of the facility, time schedule and crowdedness are the most important factors influencing satisfaction regarding facilities.

5.4 Analysis and results: Multiple regression analysis

In order to explore the effects of various neighborhood attributes on the residents’ overall assessment of their neighborhood, first, a multiple regression analysis was conducted, and then, a comprehensive analysis using “Structural equation modelling” (SEM) being applied. In general, multiple regression analysis is a popular method among researchers to estimate neighborhood satisfaction according to the changes of a set of independent variables. Such analysis can highlight the significance and strength of the relationship among multiple variables and the neighborhood satisfaction as a dependent variable. Although it can provide some invaluable insights, but is unable to capture the indirect relationships and the complex associations among the factors due to its limitations in having more than a single dependent variable. In addition, regression analysis does not have the ability to estimate the unobserved variables (which are numerous in our conceptual model). This means that the hypothetical dimensions such as social, environmental, physical, etc., which are playing key roles in our conceptualization, cannot be investigated by regression analysis.

The choice of regression analysis reflects the assumption that the scales used have interval properties, akin to many studies that utilized such scales (e.g. Diaz-

Neighborhood well-being

Serrano & Stoyanova, 2010). To examine our conceptual model by multiple regression analysis, the overall neighborhood satisfaction is treated as the dependent variable and examined by 58 independent variables. Since one of our regression assumptions is that the residuals (prediction errors) are normally distributed, therefore, checking the histogram of residuals allows us to check the extent to which the residuals are normally distributed. The well-behaved residuals can lead to correct p-values and thus are very important for any linear regression analysis. Our histogram (Figure 5.1) shows a fairly normal distribution. Thus, based on these results, the normality of residuals assumption is met.

Furthermore, the normal P-P Plot of regression residuals (Figure 5.2) can also be used to assess the assumption of normally distributed residuals. If data points (little circles) follow the diagonal line, the residuals are normally distributed. Our P-P plot shows minor deviation and thus the normality of residuals assumption is satisfied. The R-squared (R2) of the multiple regression analysis is shown in Table 5.1, which reveals that the indicators account for 73 percent of the variance in the overall neighborhood satisfaction. This suggests that variation in neighborhood satisfaction ratings are strongly related to the differences in neighborhoods and personal characteristics.

5.4.1 Interpretation

Regression coefficients in this analysis represent the magnitude of the influence of the explanatory variables on the residents’ overall neighborhood satisfaction (Table 5.2). Among all of the indicators in this analysis, Only seven variables show a significant relationship with overall neighbored satisfaction (p<.05), this might be due to the large number of variables (60) compared to the number of respondents (209). Satisfaction with “Livability”, “safety” and “image/reputation” from social domain show positive influence as expected. The other significant predictors of the overall neighborhood satisfaction are “satisfaction of road/pedestrian quality” from the “neighborhood characteristics satisfaction” category, “satisfaction of travel time to stores” from “satisfaction of neighborhood services” category, “satisfaction of size of road for biking” from the “physical activity satisfaction” category and “age” from the “socio-demographic characteristics”. The variable showing the strongest significant relationship with the overall neighborhood satisfaction is “satisfaction of safety” with the effect size of (.25). In contrast, “Age” influences the overall neighborhood satisfaction with the lowest effect size (.16). But, except “Age”, none of the socio-demographic characteristics

Neighborhood well-being

Figure 5.1 The Histogram of the overall neighborhood satisfaction

Figure 5.2 Normal Probability Plot (P-P) of the regression standardized residual dependent variable for overall neighborhood satisfaction

Neighborhood well-being

Table 5.1 Goodness-of-fit of the regression model of overall neighborhood satisfaction

R R2 Adjusted R2 Std. Error of the Estimate

.857 .734 .632 .670

Table 5.2 The estimated parameters of the neighborhood satisfaction using multiple regression analysis

Unstandardized Standardized

Coefficients Coefficients

Model Parameters B Std. Error Beta t Sig.

(Constant) 1.617 .579 2.793 .006

Age .011 .005 .162 2.189 .030

.042 Car owner .113 .159 .711 .478

-.043 Income Level -.042 .048 -.877 .382 Socio- -.104 Length of living at the house -.093 .059 -1.581 .116

demographic Household Size -.113 .070 -.121 -1.611 .109 indicators Unemployed .152 .190 .044 .798 .426

Male .022 .117 .010 .186 .853

Single person -.161 .159 -.069 -1.013 .313

Highly educated -.279 .145 -.100 -1.926 .056

Rental house -.111 .133 -.050 -.831 .407

Satisfaction of Air quality -.012 .007 -.081 -1.624 .107

Satisfaction of Houses upkeep/ .043 .048 .064 .903 .368 architecture appearance

Satisfaction of Satisfaction of Personal attachment .072 .040 .119 1.787 .076

neighborhood Satisfaction of Green space .068 .041 .107 1.636 .104

characteristic Satisfaction of Image/reputation .160 .045 .243 3.602 .000

s Satisfaction of Noise pollution .047 .037 .086 1.279 .203

Satisfaction of Cleanness .030 .044 .043 .680 .497

Satisfaction of Lighting .014 .054 .017 .258 .797

Satisfaction of Livability .099 .050 .132 2.001 .047

Satisfaction of Traffic congestion -.017 .043 -.027 -.383 .702 73

Neighborhood well-being

Satisfaction of Safety .194 .054 .248 3.622 .000

Satisfaction of Road/ pedestrian quality .098 .052 .136 1.889 .061

Satisfaction of Traffic safety (for .026 .047 .251 .548 .585 walking)

Satisfaction of Traffic safety (for .021 .057 .058 .368 .713 biking)

Satisfaction of Air quality (for -.047 .042 -.337 -1.118 .265 walking)

Satisfaction Satisfaction of Quality of path (for -.05 .059 -.610 -3.504 .001 of physical biking) activity Satisfaction of Safety/security (for -.027 .047 -.265 -.578 .564 walking)

Satisfaction of Quality of path (for .055 .042 .399 1.308 .193 walking)

Satisfaction of greenery for biking .039 .048 .114 .815 .416

Satisfaction of Size of the road .165 .070 .481 2.379 .019 (for biking)

Satisfaction of Travel time to grocery .023 .047 .056 .482 .630 store

Satisfaction of Travel time to medical .109 .055 .414 1.970 .051 center

Satisfaction of Travel time to sport -.002 .076 -.006 -.023 .982 centers Satisfaction of Satisfaction of Travel time to Services .159 .061 .568 2.607 .010 stores

Satisfaction of Travel time to park -.003 .062 -.009 -.041 .967

Satisfaction of Travel time to .086 .092 .269 .935 .351 educational centers

Satisfaction of Travel time to .043 .066 .157 .644 .520 café/restaurants

Satisfaction of Time schedule (grocery .027 .072 .063 .381 .704 store)

Neighborhood well-being

Satisfaction of Time schedule (medical .036 .073 .127 .493 .622 center)

Satisfaction of Time schedule (sport .063 .083 .230 .761 .448 centers)

Satisfaction of Time schedule (stores) -.087 .067 -.307 -1.311 .192

Satisfaction of Time schedule (park) .037 .064 .146 .580 .563

Satisfaction of Time schedule -.011 .112 -.034 -.096 .923 (educational centers)

Satisfaction of Time schedule -.063 .076 -.229 -.830 .408 (cafe /restaurant)

Satisfaction of Crowdedness (grocery -.013 .008 -.090 -1.665 .098 store)

Satisfaction of Crowdedness (medical -.093 .069 -.320 -1.353 .178 center)

Satisfaction of Crowdedness (sport .047 .069 .155 .679 .498 centers)

Satisfaction of Crowdedness (stores) -.061 .058 -.205 -1.058 .292

Satisfaction of Crowdedness (park) -.032 .068 -.118 -.469 .640

Satisfaction of Crowdedness .089 .097 .249 .921 .359 (educational centers)

Satisfaction of Crowdedness (cafe .004 .082 .013 .047 .963 /restaurant)

Satisfaction of Quality of grocery store -.021 .070 -.047 -.297 .767

Satisfaction of Quality of medical -.068 .078 -.253 -.881 .380 center

Satisfaction of Quality of sport centers -.092 .088 -.330 -1.036 .302

Satisfaction of Quality of stores -.007 .076 -.024 -.090 .928

Satisfaction of Quality of park .007 .074 .026 .095 .924

Satisfaction of Quality of educational -.139 .092 -.424 -1.513 .132 centers

Satisfaction of Quality of .039 .093 .142 .424 .672 cafe /restaurants

Neighborhood well-being shows a significant influential impact on overall neighborhood satisfaction, which is different from findings of other studies.

It is also notable that none of the attributes of the environmental satisfaction domain and physical satisfaction domain is significantly associated with overall neighborhood satisfaction, while three variables of social satisfaction domain (satisfaction of safety, livability and image) are significant in our analysis. This highlights the importance of the social domain in comparison with the other two domains in predicting overall neighborhood satisfaction. Moreover, except “satisfaction of quality of path for biking”, the signs of all other coefficients are positive, indicating that an increase in this attribute would lead to a higher level of neighborhood satisfaction, which is consistent with our framework.

Interpreting the results of regression analysis is not always straight forwards due to the internal relationships among variables. A strong relationship between an independent variable and a dependent variable could come from many other causes including the influence of other unmeasured variables. For instance, “satisfaction of size of road for biking” is not expected to have a very strong influence on the overall neighborhood satisfaction, but its high coefficient (.48) contradict it, which might be due to the strong influence of this attribute on another variable. Despite the insights we can get from multiple regression analysis, as in the beginning of this section already explained, this method has some limitations. Thus, further analysis is necessary to capture the complex mechanism of internal effects among different variables of the hypothetical model. In the next part, the same data will be analyzed using SEM method.

5.5 Analysis and results: Structural equation modelling (SEM)

The key objectives of this chapter were to evaluate and understand the interrelationships among the selected variables and neighborhood satisfaction. To achieve these objectives, SEM analysis is deemed to be a suitable method. Various research have shown that SEM is a good method in revealing the underlying structure of neighborhood as it is a hybrid model and can estimate relationships among multiple unobserved constructs (latent variables) and observable variables (e.g. Hur et al., 2010). In other words, as neighborhood satisfaction relates to complex concepts such as environmental aspects, social aspects etc. (which are difficult to be measured directly) conducting SEM should enable us to examine, address and detect all of these immeasurable constructs. A SEM model contains of two sub-models: a measurement

Neighborhood well-being model, measuring relationships among endogenous and exogenous variables (latent variables and observed variables) and the structural model, with the ability of quantifying the causal effects among the latent variables (Oud & Folmer, 2008).

5.5.1 Conceptualization

Based on the aforementioned insights from the neighborhood studies and well-being research, and the relevance of these studies to investigate the link between different neighborhood attributes, socio-demographic and household factors as well as the overall subjective assessment of the neighborhood, a conceptual framework for SEM model is designed (Figure 5.3) and will be discussed in this section.

Overall neighborhood satisfaction will be studied by exploring the relationships between the degree of neighborhood satisfaction with a set of neighborhood and personal/household characteristics. Personal/household characteristics are associated with different needs of people, while neighborhood characteristics indicate the extent by which a neighborhood offers the opportunity to satisfy these needs. Figure 5.1 presents the conceptual model of this Chapter.

Based on our survey (see chapter 3), we have three main categories of variables. The first category is the satisfaction with neighborhood characteristics, which consists of three pre-conceptualized domains: “Environmental satisfaction”, “Physical satisfaction” and “Social satisfaction”. These domains are concepts that cannot be measured directly, but linking them to other measureable variables can help quantifying them indirectly. The “Environmental satisfaction” domain is explained by four observed variables describing how residents evaluate their neighborhood environment. These four variables are the satisfaction with “Neighborhood cleanness”, “Neighborhood noise pollution”, “Neighborhood air quality” and “Neighborhood green space”. The “Physical satisfaction” domain is also identified by other four variables named as: satisfaction with “Houses upkeep & architecture”, “Road/pedestrian quality”, “Lighting” and “Traffic congestion”. Finally, the “Social satisfaction” domain consists of four attributes characterizing the grade of social satisfaction named as satisfaction with “Safety”, “Image” “Livability” and “Personal attachment”.

The second category is physical activity satisfaction, which indicates the extent to which people are satisfied with walking and biking conditions at their neighborhood. Therefore, it consists of two main domains: “Biking satisfaction” and “Walking satisfaction”. “Biking satisfaction” is evaluated by four attributes: satisfaction of “Quality

Neighborhood well-being

Overall Overall satisfaction Neighborhood Neighborhood Services Activity Neighborhood Characteristics Neighborhood

Neighborhood

Schools Cycling Walking Social Sport centers Physical Other shopping Café/Restaurants

Environmental

The conceptual modelconceptualThe

. 5 Park pedestrian pedestrian Medical centers Medical quality Personal Livability Grocery shoppingGrocery Lighting Air Quality Air Cleanness Air quality Air attachment Quality of path of Quality Quality of path of Quality

Size of the road the of Size Figure Road &Road Satisfaction of: Satisfaction crime Safety Quality greenery Landscape/ Travel timeTravel Green route Traffic safety Traffic architecture Traffic safety Traffic Crowdedness Time schedule Time Noise pollution Noise Safety regarding regarding Safety Image/reputation Traffic congestion Traffic Housing upkeep & upkeep Housing Age Socio Gender Education Marital status Marital Car ownership demographic House ownership ownership House Years of residency of Years Household income Household characteristics: And household household And Employment status Employment Number of households of Number

Neighborhood well-being of path”, “Green route”, “Traffic safety” and “Road size”. The “Walking satisfaction” is explained by satisfaction of “Air quality”, “Safety/security” and “Traffic safety”.

Finally, the third category is the “Satisfaction of services” and contains seven sub-groups: satisfaction with grocery shops, stores, restaurants/cafe, schools, medical, parks, and sport centers. All of these sub-groups are explained by four variables: satisfaction of “Travel time”, “Quality”, “Time schedule” and “Crowdedness”. However, it is worth acknowledging that conceptually, some of these attributes could belong to more than one theorized domain. So different researchers place them in different domains based on their perception and the conceptually best fit of the variables. For example, traffic congestion has been assigned to both physical and environmental domains. In addition to the entire neighborhood related variables, all socio-demographic and household characteristics (e.g. age, income, homeownership) have also been added to the model. These variables are linked to different needs of individuals and therefore may influence their neighborhood evaluation based on the extent by which different neighborhood features can fulfill their needs.

5.5.2 SEM model for neighborhood satisfaction

Our conceptual model was estimated using software package AMOS 22 with the Maximum Likelihood Estimation (MLE) method. Model was modified by excluding the relationships that were not significant at a .05 significance level. Therefore, all of the variables in the model are related to the overall neighborhood satisfaction directly and indirectly, and all paths show significant associations at the 95% confidence level. It is notable that from the set of socio-demographic factor just one variable (being single) remained in the model, and from the “satisfaction of services” category, just “satisfaction of grocery shopping” turned out to be the most significant variable. Also, “Crowdedness” variable was insignificant in explaining satisfaction with the grocery shop and thus was removed.

The goodness of the fit for the model (Table 5.3) demonstrates the fit between observed data and the hypothesized model by using several indices. One of these indices is Chi Square divided by the model degrees of freedom (/DF or CMIN/DF), which should be less than 3, and preferably close to 1 (Ullman, 2006). This value for our model equals to 2.6, which is acceptable but not the best fit. In addition, the Root Mean Square Error of Approximation (RMSEA) should be less than .08 to represent a great fit (Golob, 2003; Washington et al., 2010), and for our model it is slightly higher

Neighborhood well-being

(RMSEA=.088). Other indicators of a good fit are indices that compare target model with null model such as Comparative fit index (CFI), Incremental fit index (IFI), Tucker Lewis index (TLI), Normed fit index (NFI) and Goodness of Fit Index (GFI). If these indices exceed the amount of .90 the model tends to generate an acceptable fit. Within our initial model TLI and NFI are less than .90. Checking all of these indicators shows that our initial model needs modification to improve the relevant indices. Thus, the recommended modification from the software was considered to improve the aforementioned model quality indices.

It is worth mentioning that using the software suggested modification indices does not change the original theoretical model substantially, but adds new paths among the current variables of the models. They are double checked to be theoretically plausible especially for complex models such as neighborhood satisfaction with so many attributes. Therefore, it is decided to use supervised modification indices because of the many errors maybe assigned with the variables in a model. For our model, modification indices suggested adding some links among the latent variables of the model. Among the recommendations some were theoretically meaningful and thus they were added to the model. This resulted in creation of new paths between the latent variables of the model (i.e. social satisfaction, physical satisfaction, environmental satisfaction, walking satisfaction and biking satisfaction) as they have been ignored in the conceptual model such as linking of “satisfaction of air quality” factor with “satisfaction of biking”. Implementation of the suggested changes has led to a significant improvement in the goodness of the fit of the model as well as the R squared (R2) of the overall neighborhood satisfaction. Table 5.4 demonstrates the goodness of the fit for the improved model. It shows that Chi Square divided by the model degrees of freedom (/df or CMIN/DF) now equals to 1.76, and the RMSEA value dropped to .06, which is excellent. Also, other indicators of a good fit (CFI, IFI, TLI and NFI) have been improved significantly. Moreover, coefficients of determination or the squared multiple correlation value (R2) for the overall neighborhood satisfaction has improved from .62 to .70. So the improved model accounts for 70 percent of the total variance within the overall neighborhood satisfaction. Also, the R2 values of the other dependent endogenous variables (the overall satisfaction of each category) have been increased as for physical activity support from .29 to .43, for satisfaction of services from .37 to .48, and for characteristics satisfaction from .80 to .83. Table 5.5 presents all the direct,

Neighborhood well-being

Table 5.3 Goodness of the fit of the initial model

Fit index Estimate

Chi-square (CMIN) 719

Degrees of freedom (DF) 275

Chi-square / degrees of freedom 2.6

Root mean square error of approximation (RMSEA) .088

CFI .91

IFI .91

TLI .89

NFI .86

Table 5.4 Goodness of the fit of the improved model

Fit index Estimate

Chi-square (CMIN) 468

Degrees of freedom (DF) 267

Chi-square / degrees of freedom 1.75

Root mean square error of approximation (RMSEA) .06

CFI .96

IFI .96

TLI .95

NFI .91

indirect and total effects of the variables on the overall neighborhood satisfaction. It shows that statistically significant relationships were found among the main model variables as below:

(1) Characteristics satisfaction has a large-sized direct effect on the overall neighborhood satisfaction (.42), “Satisfaction with services” (.53) (Figure 5.4). Thereby, the “Social satisfaction” domain is a dominant construct and a key part of our model,

(2) Satisfaction of services affects the overall neighborhood satisfaction with the moderate effect (.29),

(3) Physical activity satisfaction influences satisfaction of services (.23), and have both direct and indirect effects on the overall neighborhood satisfaction (.18). 81

Neighborhood well-being

The results of the model are displayed in Tables 5.5, 5.6 and 5.7. Table 5.6 shows the relationships among the latent and observed variables participating in the measurement model. To show the reliability of each observed variable, this table includes unstandardized coefficients with p and t-value alongside standardized path coefficients and squared multiple correlations. The t-value for all the paths is bigger than 2.33. This is consistent with what Kline (1998) argued that t-value should be bigger than 1.96 for p<.05 and bigger than 2.33 for p<0, 01. Table 5.7 shows the relationships among the latent variables of the structural model.

In the following the other main results will be summarized:  Although in our initial model, neighborhood “Characteristics satisfaction” consists of “Environmental satisfaction”, “Physical satisfaction” and “Social satisfaction” but our final model indicates that only “Social satisfaction” directly influences “Characteristics satisfaction” and as a result the direct link of the other two were removed from the model, although they are affecting “Neighborhood characteristics satisfaction” indirectly (Figure 5.4).  The “Environmental satisfaction” domain influences “Physical activity satisfaction” (.51) and “Satisfaction of services” (both directly (.49) and indirectly (.12)) with large effects (Table 5.6). Also, this domain has the largest influence (.24) on the “Overall neighborhood satisfaction” after the “Social satisfaction” domain (Table 5.5).  The “Physical satisfaction” domain has indirect effects through the impact of the satisfaction of environmental domain on the “Physical activity satisfaction” (.16) and “Satisfaction of services” (.20). It can be seen that it has a moderate impact (.16) on the “Overall neighborhood satisfaction” (Table 5.5)  The “Social satisfaction” domain has direct and significant impacts on the “Environmental satisfaction” domain (.61), “Physical satisfaction” domain (.76) and “Biking satisfaction“ domain (.26) (Table 5.7). It also has a strong and direct impact on the “neighborhood characteristics satisfaction” (.90) (Table 5.6), and indirect impacts on the “Physical activity satisfaction” (.48) and “Satisfaction with services” (.53). Thereby, the “Social satisfaction” domain is a dominant construct and a key part of our model, which has a strong influence on “Overall neighborhood satisfaction” (.74).  “Walking satisfaction” and “Biking satisfaction” domains together with “Environmental satisfaction” domain influence “Physical activity satisfaction”

Neighborhood well-being

with the effects size of (.27), (.19) and (.51), respectively (Table 5.6). “Walking satisfaction” and “Biking satisfaction” also have significant and indirect effects (.07 and .04) on “Overall neighborhood satisfaction” (Table 5.5).  ”Grocery shopping satisfaction” construct is the only variable that remained in the “Satisfaction of services” category with significant associations (.17) and indirectly (.05) impacts “Overall neighborhood satisfaction” (Table 5.5) & (Figure 5.4).  Among variables of the socio-demographic group, just “Being single” remained in the model and is statistically significant with a negative sign (-.10) through mediating effects on the “Personal attachment satisfaction” (-.12), “Satisfaction of travel time to grocery shopping” (-.09), “Neighborhood characteristics satisfaction” (-.12) and “ Satisfaction of neighborhood services” (-.17).

Table 5.5 Standardized path estimates of direct, indirect and total effects

Direct Indirect Total Overall neighborhood satisfaction effect effect effect

Unobserved variables

Physical satisfaction .00 .16 .16 Social satisfaction .00 .74 .74 Environmental satisfaction .00 .24 .24 Walking satisfaction .00 07 .07 Biking satisfaction .00 .04 .04 Grocery shopping satisfaction .00 .05 .05

Observed variables

Being single .00 -.10 -.10 Physical activity satisfaction .12 .06 .18 Characteristics satisfaction .42 .00 .42 Satisfaction of services .29 .00 .29 Safety satisfaction .14 .00 .14 Road/pedestrian quality satisfaction .11 .00 .11

Neighborhood well-being

Table 5.6 Relationships among the latent and observed variables (direct effects

Squared

Latent Observed Standardized Unstandardize multiple correlation variables variables coefficient d t p s coefficient

Satisfaction of .65 .84 7.88 *** .44

Cleanness

Satisfaction of Noise .60 1.0 .39

pollution Satisfaction

Satisfaction of Green .58 .83 6.73 *** .33

space

Physical activity .51 .58 6.30 *** .43 satisfaction

Environmental Satisfaction of .49 .41 5.39 *** .48 services

Satisfaction of .75 1.01 9.21 *** .56

Road/pedestrian

quality tion

Satisfaction of Traffic .68 1.08 8.42 *** .46

atisfac congestion

Satisfaction of .63 .75 7.92 *** .39

Lighting Physical Satisfaction of Houses .68 1.00 .46 upkeep/ architecture

Satisfaction of .69 1.48 7.34 *** .49

Personal attachment

Satisfaction of .66 1.32 7.84 *** .44 Image/reputation

Satisfaction of .65 1.13 7.62 *** .42 satisfaction Livability

Social Satisfaction of Safety .60 1.00 ** .36

Characteristics .90 1.31 9.38 *** .83 satisfaction

Satisfaction of Road .98 1.09 37.04 *** .96

size

Biking Satisfaction of Quality .97 1.10 34.12 *** .93

satisfaction of path

Neighborhood well-being

Satisfaction of Air .14 .37 2.80 ** .54 quality

Satisfaction of Traffic .95 1.00 .90

safety

Satisfaction of Green .93 1.04 29.61 *** .87 route

Physical activity .19 .09 3.42 *** .43 satisfaction

Satisfaction of Traffic 1.04 1.00 1.07

safety

Satisfaction of .96 .92 37.24 *** .92

Safety/security satisfaction Satisfaction of Air .72 .50 14.04 *** .54 quality

Walking Physical activity .27 .03 4.90 *** .43 satisfaction

Satisfaction of Time .98 1.00 .97

schedule 1.00

Satisfaction of Quality .95 .93 33.87 *** .90

Satisfaction of Travel .89 .95 25.67 *** .80

time satisfaction

Groceryshopping Satisfaction of .17 .07 3.27 *** .48 services

*** Significant at.001 level, ** significant at.01 level.

Table 5.7 Relationships among the latent variables

Standardized Unstandardized coefficients Endogenous Latent coefficient

variables t p

Physical satisfaction .76 1.03 6.88 *** Social Environmental satisfaction .61 .87 4.48 *** satisfaction Biking satisfaction .26 .91 3.44 ***

Environmental satisfaction .32 .34 2.67 ** Physical

satisfaction Physical satisfaction .12 .01 2.47 ** Walking

satisfaction *** Significant at.001 level, ** significant at.01 level.

Neighborhood well-being

for neighborhood satisfaction forneighborhood

Results of SEM of the modified model with standardized coefficients standardized coefficients with the of modified model of SEM Results

Figure

Neighborhood well-being

5.5.3 Interpretation and discussion

Generally speaking, our results support the main parts of the initial conceptual model and are also coherent with the previous studies. Although some variables are insignificant and have been omitted from the model, but all principal categories and key aspects have remained and show significant influence on overall neighborhood satisfaction. Using suggested modification indices led to significant improvements as well as getting new insights about our data set. Below are the main finding and interpretations from the results:  As expected, satisfaction with neighborhood characteristics plays an important role on the overall satisfaction with neighborhood, since it consists of the most critical factors affecting the feeling toward living in an area.  Satisfactions with social, environmental and physical conditions of the neighborhood are the most important aspects, which have also been admitted by several studies (e.g. Hur & Jones, 2008).  Satisfaction of social dimension directly and indirectly influences all the three major categories of neighborhood characteristics satisfaction, physical activity satisfaction and satisfaction with services.  The more satisfied with the environmental situation of the neighborhood, the more likely respondents were to be satisfied with the physical activity and services of the neighborhood.  The impact of physical dimension satisfaction on the overall neighborhood satisfaction is indirect and is through the impact on the environmental dimension satisfaction, which makes sense as the physical and environmental features of neighborhoods are highly associated and to some extend are tied together.  The dominant influence of social satisfaction domain on the overall neighborhood satisfaction basically reflects that residents who reported stronger levels of social satisfaction also tend to have higher level of neighborhood satisfaction. Therefore, poor neighborhood is strongly associated with poor social conditions and specially the lack of safety. This means that improving physical features and conditions will enhance residents’ satisfaction only if the social conditions are developed.  The physical activity satisfaction accounts for 19% of the variance of the overall neighborhood satisfaction both directly and indirectly through the

Neighborhood well-being

impact on satisfaction of services. The indirect influence clearly indicates that some residents tend to walk or cycle for their daily needs such as grocery shopping. This furthermore infers that better environment and infrastructure for active lifestyle can clearly provide easier shopping access. Especially for senior residents, it is proved that the ease of access and walkability within the area influences the service use.  For our respondents, the impact of satisfaction with environmental condition is much higher than satisfaction with walking and biking infrastructure to assess neighborhood activity satisfaction. This can be interpreted that the satisfaction of environmental qualities such as level of noise, greenery and cleanness are significant predictors for doing physical activity. Clearly, lacking of a good environment reduces motivation for doing outdoor activities even if the physical infrastructure were sufficiently provided.  Satisfaction with the social environment also associates with biking satisfaction. This makes sense as doing physical activity relates to the social features such as feeling safe (Van Cauwenberg et al., 2018). It is also well proved that fear of crime reduce desire to do physical activity in urban area.  Neighborhood characteristics satisfaction on one hand, explained 42% of the variance in the neighborhood overall satisfaction and on the other hand, it is mainly associated with the social condition satisfaction (.90) and has no significant relationship with physical and environmental condition satisfaction. This indicates that for the residents, neighborhood characteristic are almost equal to their social conditions, which is different from our initial conceptualization.  The direct impact of the safety satisfaction attribute on the overall neighborhood satisfaction describes the importance of the residents’ expressed evaluation of the safety in predicting neighborhood satisfaction. Many previous studies had also confirmed our findings and indicated that lack of safety is associated with social isolation, limited freedom and also deprivation in neighborhoods (e.g. Visser et al., 2015).  Grocery shopping satisfaction is the only significant service activity remaining in our model indicating that other facilities (medical centers, et.) should not necessarily be located in the same neighborhood. That is because of the widely uses of grocery shops by everyone in a neighborhood and the limited or no use of other facilities in everyone’s daily life. Satisfaction of services is significantly 88

Neighborhood well-being

influenced by the environmental satisfaction of the neighborhood (.61). This can highlight the fact that the external conditions of services are dominant in comparison with internal factors such as quality, time schedule and crowdedness of the services for our respondents. This might be due to the availability of chain supermarkets with the same quality and time schedule in most of the areas that can lower the impacts of the internal factors.  Social conditions satisfaction also indirectly influences satisfaction of services (.53) through the impacts on the physical activity satisfaction. This again refers to the importance of external conditions such as ease of access and safety evaluation. Especially, service use for senior residents is dependent on the ease of access and walkability of the area.

5.6 Conclusions and discussion

This chapter presents a methodological and empirical contribution to the literature on the residents’ satisfaction of neighborhood using SEM method. SEM allows analyzing a multiple layers of information about neighborhood and the links among them simultaneously. It also enables us to have a better understanding of the complexity of the neighborhood concept. Our findings highlight the major and distinct influence of the social features on the residents’ overall satisfaction regarding neighborhood. This indicates that emphasize should be placed on improving social domain within neighborhoods.

The context of the research is very important in neighborhood studies as different countries may have different policies and strategies for neighborhood design and spatial land use. Considering the Netherlands as our data collection base point, it is worth mentioning that the Dutch approach in neighborhood design and planning is comprehensive, and addresses all aspects of living. It is based on the mixed and compact land use concept, which provides proper access to public transport and necessary infrastructure in multifunctional neighborhoods. So basically, the physical components of the surveyed neighborhoods can fulfill residents’ main needs, while concerns over social aspects are remained. This can be mainly explained by the needs hierarchy theory, which describes that once the individuals’ lower needs have been fulfilled, they seek to fulfill higher levels of needs.

Although this chapter has examined all the key concepts of neighborhood satisfaction, considering the numerous factors and domains reflecting neighborhood

Neighborhood well-being satisfaction on one hand, and the dynamic nature of neighborhood on the other hand, there are many opportunities to expand research in this field in future studies.

Transport well-being

6.1 Introduction

Following our domain-specific satisfaction approach, including satisfaction with housing, neighborhood and job, travel satisfaction is another potentially important domain- specific satisfaction, which may affect well-being (Schimmack, 2008; Ettema et al., 2010). Transport is an inevitable and inseparable part of urban life. It provides access to all necessary locations and facilitates out of home activities, and enables people to participate in certain experiences such as employment, education, social and leisure activities. A good transport system can influence different components of everyday living positively, whereas inefficient transport can be costly, time consuming, risky and stressful for commuters, and results in a huge loss of abilities and resources. This accordingly contributes to substantial diminution of individual and public overall life satisfaction (e.g. Cao, 2013). However, despite the important, tangible and

Transport well-being consequential implications of travel experiences for well-being, the topic has just recently received attention, and limited number of studies focused on the link between travel satisfaction and well-being (e.g. Sirgy et al., 2011; Ettema et al., 2011; Friman et al., 2013; Aydin et al., 2015; Gao et al. 2017). This is partly because of the broad scope and various facets of this link, which makes researchers to normally focus and examine the effects of an explicit component of travel satisfaction on well-being rather than the entire travel satisfaction.

In formulating the conceptual framework of this thesis (see Chapter 3), transportation contributes to quality of life, and thus, this chapter will focus on the transportation domain-specific subjective well-being. To assess the quality of life associated with this domain, judgments of the individuals’ satisfaction with travel will be addressed and explicitly, the following research questions will be explored:

I. What are the main predictors of travel satisfaction? Or which of the travel features have received empirical support in predicting travel satisfaction?

II. To what extent do these indicators contribute to predicting travel satisfaction?

III. What are the relationships among these indicators and to what extent do they affect each other?

6.2 Research background

Travel satisfaction is an emerging topic in travel behavior science and as a result the literature is still limited but rapidly expanding (Gatersleben & Uzzell, 2007; Friman & Fellesson, 2009; Ettema et al., 2012; Cao, 2013; Olsson et al., 2013; Susilo & Cats, 2014; Mokhtarian & Pendyala, 2018). Some studies have included built environment and psychological dispositions (e.g. attitude and personality) in evaluating commuters’ satisfaction (e.g. Friman et al., 2013; St-Louis et al., 2014; De Vos et al., 2016). In general, the concept of commute satisfaction mainly comes from customer satisfaction research, where customers’ reaction to the offered services can be considered as customer satisfaction Moreover, an approach called Satisfaction with Travel Scale (STS) has been developed based on the cognitive and affective judgments of satisfaction and has been used widely in travel satisfaction studies (e.g. Friman et al., 2013).

Transport well-being

6.3 Factors determining travel satisfaction

In addition to socio-demographic of individuals, their access and use of travel modes and their trip attributes influence their overall judgments regarding the travel. A growing number of empirical studies have examined the role of instrumental factors such as travel time, travel cost and level of service (e.g. Friman & Fellesson, 2009; Ettema et al., 2012; Cao, 2013; Olsson et al., 2013; Susilo & Cats, 2014) or non- instrumental variables such as safety, comfort and social interactions (e.g. Stradling et al., 2007) on travelers’ satisfaction. Along the same line, a vast majority of current studies have explored the mode-related travel satisfaction (e.g. Páez & Whalen, 2010; Friman et al., 2013; St-Louis et al., 2014). Furthermore, some recent studies also mentioned that travel satisfaction varies among urban areas, and thereby the role of built environment and neighborhood on the travel satisfaction has been explored (e.g. Cao & Ettema, 2014; De Vos et al., 2016). Moreover, few studies explored the impact of travelling in rush hours and the negative influences of unpredictability and stress on the overall trip satisfaction (e.g. Morris & Hirsch, 2016). This implies that the determinants of travel satisfaction are numerous, and in the following, we will explain them in more detail.

6.3.1 Individuals’ socio-demographic and household characteristics

Socio-demographic variables are linked with the individuals’ needs, desires, time and money constraints, and thereby potentially influence travel satisfaction (e.g. Cao et al., 2015; De Vos et al., 2016, Higgins et al., 2017). The main socio-demographic variables that have been discussed in the literature are age, household income, employment, household composition, gender and car ownership. There are few studies that investigate the role of individuals’ health on travel satisfaction, and it has been investigated among the socio-demographic variables (e.g. Ye & Titheridge, 2016). However, empirical studies provided mixed findings when it comes to satisfaction variations across socio-demographic variables.

In this regards, age probably is the most discussed variables in the literature. Many studies found that travel satisfaction increases linearly with age (e.g. Cao and Ettema, 2014; De Vos et al., 2016), but a few other studies revealed that the relationship is U- shaped, means younger adults and seniors experience higher satisfactions of commuting in comparison with middle aged commuters. Adversely,

Transport well-being some authors argued that increase in age contributes to difficulty in mobility for some people (specially the lower incomes), which leads to lower travel satisfaction. Also, car ownership is likely to influence travel mode choice and accessibility, which consequently influence travel satisfaction (Ding et al., 2017). Regarding gender, some researchers observed gender differences in travel satisfaction. The results are mixed as some found male are more satisfied (e.g. Higgins et al., 2017), while some other found evidence that males tend to be less satisfied compared to females (e.g. Gao et al., 2017).

6.3.2 The built environment

Transport is context dependent and thus, the satisfaction people derive from their daily trips is very much influenced by the physical environment surrounding them (Clark et al., 2016). To date, a growing number of studies have linked built environment with travel satisfaction (e.g. Friman et al., 2013; Cao & Ettema, 2014; De Vos et al., 2016; Ye & Titheridge, 2017). However, since the topic is very broad, to achieve a sound understanding, different researchers have focused on one or few aspects of the physical space. Based on the research findings, residential location, access to different transport modes, density and land use are the most important characteristics of the built environment that can highly contribute to travel satisfaction. As most trips start from home, residential location affects many aspects of commuting. House location within the neighborhood and neighborhood location within the city, all together defines residential location, which is distinctly linked to the accessibility notion. As Lynch (1984) described, accessibility is the main functional character of urban spaces, contributes to the ability of residents in participating in different activities and using different resources and information. At a micro-scale, a good neighborhood can provide access to main facilities, such as supermarkets, kids’ playgrounds and public transport means. A neighborhood that is easily accessible on foot is preferred by everyone especially elderly residents and other vulnerable groups. At a wider scale, the whole area that consists of different blocks and small neighborhoods should provide easy access to health care, education, employment, recreation and other key services.

Commuting mode choice is another key player in travel satisfaction, which is directly influenced by the physical built environment (e.g. Cervero, 2002; Chen et al., 2008; Munshi, 2016). Mode choices generally impact the perception of an area and

Transport well-being connectivity feeling. As a result, residents of suburban area are likely to have less transport satisfaction in compared with inner city residents, mainly because of the mode choice limitations (e.g. De Vos et al., 2016). Dense urban areas and mixed land use (mix of residential and services) which are “New urbanism” planning paradigms contribute to shorter distances and travel time, which consequently can enhance travel satisfaction. In addition to that, walking and biking, which are proven to be the most satisfying travel modes, are promoted in dense urban areas. Friman et al. (2011, 2013) found that living close to work and school contribute to higher level of satisfaction since the likelihood of walking or cycling increase. Furthermore, those who looks at travel time as a buffer between multiple activities believes that active travel modes use is the most relaxing activity, reducing stress and increasing the travel enjoyment (e.g. Urry, 2016).

Aside from short distance that dense urban areas can provide, urban infrastructures such as availability of good pedestrian passes and separated bike routes are important facilitators for active traveling. There are also some other factors that indirectly encourage active travel mode use such as narrow streets, traffic congestion, and lack of parking spaces which are physical barriers, and can reduce car commuters satisfaction (e.g. Maat & Timmermans, 2009). In general, neighborhood satisfaction directly influence travel satisfaction as people who live in unfavorable areas experience lower travel satisfaction due to restrictions on their travel mode choice (De Vos et al., 2016).

6.3.3 Travel mode choice

Previous research have linked travel mode choice to travel satisfaction and tried to find out which mode users are the most satisfied commuters. The results, however, show some agreements and disagreements when discussing satisfaction of commute by different modes. In terms of active travel modes, most of studies have found that cyclists and walking people tend to be the most satisfied (e.g. Gatersleben & Uzzell, 2007; Páez & Whalen, 2010; De Vos et al., 2016), and particularly cyclists have the highest level of satisfaction (e.g. Avila-Palencia et al., 2018). In contrast, public transport commuters display the lowest satisfaction compared to the users of other modes (Páez & Whalen, 2010; Friman et al., 2013). Furthermore, some researchers found that car commuters are more satisfied with their travel than public transit users

Transport well-being

(e.g. Turcotte, 2005), while the others found higher satisfaction among public transit than that of private transportation (e.g. St-Louise et al., 2014; Olsen et al., 2017). In general, people choose different transportation mode based on their needs, desires and their time-space constrains which is related to socio-demographic characteristics and their personality.

6.3.4 Time, the role of rush hours

Time in general, is a main notion in the dynamic world and performing activities are bonded to space, time or both. The fixed scheduled activities are bonded to both space and time, while the flexible activities can be rescheduled and shifted in time and space. Normally, people tend to shift their flexible activities to avoid rush hours and improve their commute experience (Schwanen et al., 2008; Oakil et al., 2016). Traveling in rush hours can significantly influence travel satisfactory as it may affect all the main trip attributes (i.e. Travel time, travel cost, travel safety and travel comfort). It is likely that trips in rush hours prolonged due to traffic congestion and crowded roads, and thereby the overall travel satisfaction can negatively be influenced during rush hours (Morris & Hirsch, 2016). Also cost of trips could increase significantly during rush hours. In many countries, rush hours fees for public transportation, especially for the rail means are higher. Moreover, because of traffic congestion, vehicles consume more fuel and the possibility of finding free parking spaces is less. This makes rush hours trips more expensive for both public and private transportation commuters. Travel safety in rush hours is likely to be reduced, as the probability of accidents is getting higher in busy roads. Additionally, delays and extreme on-board crowding in public means, and stress exposed by traffic congestion to car commuters can reduce travel comfort significantly during rush hours.

6.3.5 Trip attributes satisfaction

Travel satisfaction has been commonly conceptualized as a function of trip attributes (Gao et al., 2016). The previous studies consider numerous travel-related attributes in order to estimate travel satisfaction. Travel time, travel cost, travel safety and travel comfort are the most examined variables (e.g. Beira & Cabral, 2007; Aydin et al., 2015). In other words, the extent to which commuters feel happy with the trip attributes strongly contributes to the overall evaluation of their trips, so it is logical to study them distinctly in the following.

Transport well-being

6.3.5.1 Travel time (trip duration) satisfaction

Time is very valuable and time saving is always a major benefit of a great transportation system. Generally, travel time is the entire time devoted to travelling. This includes delays, transfer time, waiting in traffic congestion, etc., and can be influenced by many aspects such as vehicle’s speed and land use mix (St-Louis et al., 2014). Shorter travel time is usually preferred, especially for frequent journeys such as daily trips to work, school etc. (e.g. Morris & Hirsch, 2016). Therefore, clearly the frequencies of trips together with the length of trips are the main determinants of travel time satisfaction. Zhang et al. (2016) examined travel time and frequency of public transport as the predictors of travel satisfaction and found that they significantly influence travel satisfaction. Several other studies have also found negative associations between longer trips and commuter’s satisfaction, and related it to reasons such as stress level, exhaustion and traffic congestion (e.g. Morris & Guerra, 2015). For instance, Higgins et al. (2017) found that longer travel time reduces travel satisfaction among the Canadian commuters, specifically when considering the travelers’ perception of congestion.

6.3.5.2 Travel cost satisfaction

Satisfaction with travel cost is another key attributes of travel satisfaction. Clearly, travel needs allocation of proper budget and travel cost is defined as all of the expenses associated with a trip. As costs and benefits are always related, satisfaction with travel cost is substantially linked with other trip attributes (i.e. travel time, travel comfort and travel safety),as well as the purpose of trip and activity at the destination. According to “Microeconomic theory”, individuals tend to minimize their travel cost unless they receive compensation for it (Stutzer & Frey, 2008). This implies that affordability or satisfaction with travel cost is an important aspect not just for travel satisfaction, but also at a higher level for individuals to participate in society and non-paid activities. Especially, for low-income people high-cost transportation is a burden and leads to isolation and lower well-being (e.g. Bocarejo & Oviedo, 2012).

6.3.5.3 Travel safety satisfaction

Safety is an important human need and thus the extent to which commuters feel safe during their trips is a key initiative in travel satisfaction. Two major domains usually 97

Transport well-being explain travel safety: safety from accidents and safety from crime (Joewono & Kubota, 2006). Whilst, the former refers to components such as roads safety, vehicles safety, driving qualifications, etc., the latter is a reflection of safety in community and rates of criminal incidents. Typically, private transportation is assumed to impose less risk regarding crime, whereas public transit is seen safer in terms of accidents (especially for road trips). Iseki and Smart (2012) found satisfaction levels for safety to be more associated with the overall travel satisfaction than other attributes (e.g. reliability, amenities etc.) among transit users. Also, Spears et al. (2013) argued that personal safety concerns significantly associated with commuters' satisfaction. For active commuters (cyclists and walking people) both crime and accident could potentially be a threat based on the context. Many authors argue that improving safety features could motivate people to switch their travel modes to active modes, which in turn increase travel satisfaction (e.g. Ettema et al., 2016). The safety assessment is associated with both direct and indirect experiences of individuals, and therefore, varies among commuters. It is also reported that safety feeling varies among gender and different age groups (Susilo & Cats, 2014).

6.3.5.4 Travel comfort (convenience) satisfaction

The last travel attribute that influence travel satisfaction is travel comfort satisfaction. In most of travel satisfaction studies, travel comfort has been considered as a determinant component of a trip (e.g. Stradling et al., 2007; Aydin et al., 2015). For instance, Fellesson and Friman (2008) examined the perceived satisfaction with public transport in nine European cities, and found that the most important factors influencing travel satisfaction are comfort together with safety, staff and system dimensions. Convenience and comfort are the two concepts that cover a wide range of aspects and feelings and usually used interchangeably to explain the ease of use and the amount of services provided to make a pleasant trip. In general, travel comfort satisfaction is related to both physiological and psychological conditions. In terms of physiological comfort ability, vehicle’s standards and thermal situation are among the key factors to have a pleasant feeling. For public means, in addition to the above- mentioned aspects, distance to station and number of passengers (vehicle’s crowding), and cleanness are among the key determinant factors (e.g. Dell’olio et al., 2011; Redman et al., 2013). Feeling stressed, freedom, social interactions and possibility of

Transport well-being doing various activities in long journey are among the factors that impact psychological comfort feeling. Stress is the most psychological threat that is discussed for comfort feeling and travel satisfaction. It is said that drivers tend to experience stress more often, while journeys by public transport are more stress free (Novaco & Gonzales, 2009; Legrain et al., 2015; Mattisson et al., 2016). However, public transport’s unpredictable delays can also trigger feeling stressed during journeys (e.g. Cats & Gkioulou, 2017). Independence and freedom feeling are the other two main aspects that influence psychological comfort (Steg, 2005). Private modes can makes journeys more convenient, but on the other hand, can impose isolation in longer single occupancy journeys (Mattisson et al., 2015). However, during the trips by public transport modes, the possibility for interactions and social contacts with passengers and/or driver can change the trip feeling (Bissell, 2010). Stradling et al. (2007) found that variables such as comfort, stress, social interaction and scenery have a great influence on trip satisfaction. In addition to all of the aforementioned factors, the possibility of committing various types of activities (such as reading, watching movie, etc.) in journeys by public means is likely to increase the comfort feeling for many commuters (Ettema et al., 2012).

6.4 Analysis and results: Multiple regression analysis

In order to explore the implicit effect of various travel components on the commuters’ overall assessment of their travel, a multiple regression analysis was conducted with overall travel satisfaction as the dependent variable and 45 independent variables (including socio-demographic variables). We assumed that the scales, which used to measure satisfaction have interval properties (same degree of difference between items) based on many studies that utilized such scales (e.g. Diaz-Serrano & Stoyanova, 2010). It is notable that those who walk, bike and not travel at all have not evaluated anything regarding their trip characteristics. Therefore, a binary variable called mode choice has defined which contains “Traveler” (including public or private transport mode user), and “Non-traveler” (including walk, bike or not travel at all) to represent this group in our model. Apart from the socio- demographic variables that are binary (refer to Table 3.3), age, household income and household size were treated as continuous variables. Also, “Health” included as a socio-demographic variable and is a self-reported health of individuals (rating).

Transport well-being

Furthermore, the normal P-P Plot of regression residuals (Figure 6.2) can also be used to assess the assumption of normally distributed residuals. If data points (little circles) follow the diagonal line, the residuals are normally distributed. Our P-P plot shows minor deviation and thus the normality of residuals assumption is satisfied. The results about R-squared (R2) and adjusted R-squared are shown in Table 6.1. It shows that the indicators accounted for almost 46 percent of variance of the overall travel satisfaction.

6.4.1 Interpretation

Regression coefficients in this analysis represent the magnitude of the influence of the explanatory variables on the residents’ overall travel satisfaction (Table 6.2). Among all of the indicators in this analysis, only two variables show a significant relationship with the overall travel satisfaction (p<.05). This might be due to relatively small sample size compared to the number of variables in the model. “Overall neighborhood satisfaction” (.34) and "Travel time to family /friends” (.17) from accessibility satisfaction category both significantly influence “Overall travel satisfaction”. Clearly and according to literature (e.g. Ewing & Cervero, 2010) neighborhood satisfaction highly influences travel satisfaction as it determines distances to different places and access to different modes of transport. Satisfaction with "Travel time to family/friends” has been also mentioned in other travel satisfaction studies as an influential variable (e.g. St-Louise et al., 2014; Ye & Titheridge, 2016; Higgins et al., 2017).

Although multiple regression analysis provides some insights into the impact of socio-demographic characteristics and all the mentioned built environment and travel- related characteristics on overall travel satisfaction, but this method is unable to capture the indirect relationships and the complex associations among these variables. Thereby, further analysis is necessary to understand the mechanism of effects among the study variables, which will be explained in the next part.

100

Transport well-being

Figure 6.1 The Histogram of the overall travel satisfaction

Figure 6.2 Normal Probability Plot (P-P) of the regression standardized residual dependent variable for overall travel satisfaction

101

Transport well-being

Table 6.1 Goodness-of-fit of the regression model of overall travel satisfaction

R R2 Adjusted R2 Std. Error of the Estimate

.677 .458 .307 .952

Table 6.2 The estimated parameters of the transport satisfaction using multiple regression analysis

Unstandardized Standardized

Coefficients Coefficients

Model Parameters B Std. Error Beta t Sig.

(Constant) 3.103 .820 3.783 .000

Age -.001 .005 -.008 -.102 .919

Household size -.052 .120 -.054 -.436 .663

Unemployed -.334 .263 -.094 -1.273 .205

Socio- Household with kids .160 .236 .065 .680 .497 demographic Highly educated .147 .196 .051 .753 .453 indicators Male .245 .157 .107 1.553 .122

Single -.301 .228 -.124 -1.322 .188

Car owner .021 .244 .008 .087 .931

Health .087 .048 .125 1.800 .074

Household income .006 .051 .008 .113 .911

Travel cost (private transport) -.044 .100 -.124 -.434 .665

Travel cost (public transport) .144 .184 .266 .782 .435

Crowdedness (public transport) .068 .199 .123 .340 .734

Travel time (private transport) .053 .136 .146 .388 .698

Travel time (public transport) .195 .243 .385 .802 .424

Peak hours Parking availability .102 .090 .305 1.124 .263

satisfaction Parking cost -.033 .063 -.084 -.518 .605

Safety (public transport) -.316 .392 -.675 -.806 .422

Traffic congestion .091 .136 .233 .673 .502 (private transport) 102

Transport well-being

Waiting time (public transport) .106 .319 .231 .333 .740

Safety (private transport) .090 .119 .289 .760 .448

Distance to station-stop -.206 .318 -.484 -.647 .518

Inside public means quality .139 .296 .281 .469 .640

Travel cost (public transport) .344 .219 .760 1.565 .119

Travel cost (private transport) -.114 .114 -.372 -1.004 .317

Crowdedness (public transport) .344 .219 .760 1.565 .119

Travel time (public transport) .331 .292 .760 1.135 .258

Travel time (private transport) -.160 .133 -.537 -1.202 .231 Off-peak Parking availability .017 .078 .055 .223 .824 hours Parking cost .106 .062 .287 1.728 .086 satisfaction Safety (public transport) -.210 .247 -.507 -.849 .397

Traffic congestion (private .067 .121 .209 .551 .582 transport)

Waiting time (public transport) -.333 .212 -.756 -1.570 .118

Safety (private transport) .090 .119 .289 .760 .448

Distance to station/stop .064 .201 .167 .316 .752

Inside public means quality .049 .228 .114 .216 .829

Travel time to work -.033 .022 -.114 -1.503 .135

Accessibility Travel time to kids school/day -.002 .027 -.007 -.077 .939 satisfaction care

Travel time to service places .028 .020 .104 1.390 .166

Travel time to family/friends .064 .032 .166 2.043 .043

Travel time to recreational places -.003 .024 -.010 -.127 .899

Overall neighborhood .348 .097 .335 3.583 .000 Built Neighborhood infrastructure for environment .052 .064 .061 .813 .417 walking/biking satisfaction Neighborhood safety .013 .063 .016 .203 .839

Travel Non-traveler .805 .484 .345 1.664 .098

103

Transport well-being

6.5 Analysis and results: Structural equation modelling (SEM)

The key objectives of this chapter are to evaluate the proposed theoretical model and to understand the interrelationships among the study variables in formulating travel satisfaction. Structural equation modeling (SEM) is deemed to be an adequate solution for the achievement of these objectives and has been increasingly used in studying the travel satisfaction (e.g. de Oña et al., 2016; Wan et al., 2016; Friman et al, 2017; Gao et al., 2017). The recently published studies on travel satisfaction have shown that SEM is a good method in estimating the relationships among multiple unobserved constructs (latent variables) and observable variables, and increases the statistical efficiency.

6.5.1 Conceptualization

Based on the aforementioned insights from different travel studies and the relevance of these studies to investigate the link between different travel attributes, built environment variables and socio-demographic variables as well as the overall travel satisfaction, a conceptual model is elaborated. Figure 6-3 depicts the hypothetical model structure that we estimated to examine the influence of different attributes on overall travel satisfaction. As the figure shows, overall travel satisfaction is assumed to be influenced by travel mode choice in peak and off-peak hours and the satisfaction with the associated trip attributes, which predicts public and private transport satisfaction in peak and off-peak hours. In addition to the trip related factors, satisfaction with the built environment factors may influence overall travel satisfaction. These variables include neighborhood satisfaction and its related variables, and also feeling regarding the accessibility to important places in daily life. Depending on socio- demographic variables, people may choose particular transport means and may differently rate their satisfaction with travel. Therefore, the socio-demographic variables may directly and/or indirectly influence overall travel satisfaction.

In our conceptual model, five variables are incorporated as latent variables, these variables are complex concepts, and thereby linking them to measurable variables can help quantifying them. “Peak hours private transport satisfaction”, “Peak hours public transport satisfaction”, “Off-Peak hours private transport satisfaction”, “Off-peak hours public transport satisfaction” and “Accessibility” are the five pre-conceptualized constructs. Both “Peak hours private transport satisfaction” and “Off-peak hours private transport satisfaction” are measured by five main attributes. They consist of satisfaction with: “Travel time”, “Travel cost”, "Travel safety”, "Parking availability”, "Parking cost”

104

Transport well-being and “Traffic congestion”. Also, both “Peak hours public transport satisfaction” and “Off- Peak hours public transport satisfaction” are measured by seven variables. They which consist of satisfaction with: “Travel time”, “Travel cost”, Travel safety”, “Inside means quality”, “Crowdedness”, “Waiting time” and “distance to station/stop”. “Accessibility” is explained by 5 attributes describing the extent to which residents feel that different places are accessible and rating their satisfaction with: “Travel time to work”, “Travel time to family/friends”, “Travel time to recreational places”, “Travel time to services places”, and “Travel time to kids’ school/daycare”.

6.5.2 SEM model for transport satisfaction

Our conceptual model was estimated using AMOS 22 software package with the Maximum Likelihood Estimation (MLE) method. This method assumes multivariate normality, which enables estimating the direct and indirect correlations among factors, and the p values of the paths including the standardized regression weights of the conceptual model. In the first step, all of the attributes were included in the model to see whether they have any direct or indirect relation with the observed and unobserved endogenous variables. After omitting the insignificant relationships from the model, most of the socio-demographic variables are remaining in the model (except “Household size”, “Unemployed” and “Highly educated”). Also, both “Off-Peak hours Public transport satisfaction” and “Off-Peak hours private transport satisfaction” were found insignificant, and thus, are excluded from the model. The final model consists of 30 observed variables and 3 unobserved endogenous variables (“Peak hours public transport satisfaction”, “Peak hours private transport satisfaction” and satisfaction with “Accessibility”).

The results of the final SEM model (Table 6.3) demonstrate that the model excellently fit to the data. Chi Square divided by the model degrees of freedom (/df or CMIN/DF) is an essential measure and is equal to 2.2 for our model. In addition to that, the Root Mean Square Error of Approximation (RMSEA) should be less than .08 to represent a great fit (Golob, 2003; Washington et al., 2010) and in our model RMSEA value is .07. Other indicators of a good fit as mentioned by many researchers (e.g. Moss, 2009) are indices that compare target model with null model, such as Comparative fit index (CFI), Incremental fit index (IFI), Tucker Lewis index (TLI), Normed fit index (NFI). If these indexes exceed the amount of .90 the model tends to

105

Transport well-being

- Socio household household nd demographic characteristics a : peak) - centers satisfaction satisfaction family/friends quality places Travel Cost Travel Travel time Travel Travel Cost Travel Parking cost Parking Travel time Travel Travel Safety Safety Travel Stop/station Stop/station Travel Safety Travel Waiting time Waiting Crowdedness Insidemeans Travel time to Travel school/day care school/day Traffic congestion Traffic distance to to home distance Parking availability Parking Travel time to timeto kids’ Travel Travel time to timeto work Travel recreational recreational Travel time to timeto service Travel Travel to to Travel Peak and Off and Peak (

Trip attributes attributes Trip

transport transport transport rivate rivate

Peak hours hours Peak model p - Peak hours hours Peak private private

transport transport transport transport Peak hours Peak satisfaction satisfaction satisfaction Peak hours Peak satisfaction Off Off public public

Accessibility

The conceptual The

. 6 Travel

Overall Overall

Satisfaction Figure Overall Overall neighborhood : peak) - - for for safety Mode choiceMode environment satisfaction: environment Non Walk Bike Public Public traveler Private transport transport infrastructure infrastructure Neighborhood Neighborhood biking/walking Built Travel Travel (Peak and Off and (Peak

106

Transport well-being

Table 6.3 Goodness of the fit of the model

Fit index Estimate

Chi-square (CMIN) 777

Degrees of freedom (DF) 354

Chi-square / degrees of freedom 2.19

Root mean square error of approximation (RMSEA) .07 CFI .95

IFI .95 TLI .94

NFI .91

generate an acceptable fit. Checking all of these indicators shows that our model perfectly fits. Moreover, R squared (R2) of the ultimate endogenous variable (called squared multiple correlation value in AMOS) equals to .45, which means our proposed theoretical model satisfactorily accounts for 45 percent of the variance in residents’ overall travel satisfaction. Table 6.4 presents all the direct, indirect and total effects of the variables on the overall travel satisfaction. It shows that statistically significant relationships were found among the main model variables as below:

 “Non-traveler” in peak hours has the largest effect on “overall travel satisfaction” (.42) followed by “Overall neighborhood satisfaction” (.39).

 Both “Peak hours private transport satisfaction” (.27) and “Peak hours public transport satisfaction” (.15) positively influence “overall travel satisfaction”.

 “Accessibility” also has a direct significant relationship with “overall travel satisfaction” (.13).

 Satisfaction with neighborhood safety and neighborhood physical activity both are associated with neighborhood satisfaction and through that contribute to overall travel satisfaction.

 “Health” is the only variable from the socio-demographic category that directly correlates with travel satisfaction (.10). It is also the most influential variable among the socio-demographic variables with total effect size (.26).

107

Transport well-being

 In addition to the direct effects on travel satisfaction, neighborhood satisfaction also influence travel satisfaction indirectly by affecting travel attributes such as satisfaction with "Travel time" and "Parking cost" (Figure 6.4).

 “Male“ and “Income level” both have negative effect on overall travel satisfaction.

Table 6.5 shows the relationships among the latent and observed variables participating in the measurement model. To show the reliability of each observed variable, this table includes unstandardized coefficients with p and t-value alongside standardized path coefficients and squared multiple correlations. The t-value for all the paths is bigger than 2.33. This is consistent with what Kline (1998) argued that t-value should be

Table 6.4 Standardized path estimates of direct, indirect and total effects on overall travel satisfaction

Direct Indirect Total Overall transport effect effect effect

Unobserved variables

Accessibility satisfaction .13 .0 .13 Peak hours public transport satisfaction .15 .0 .15

Peak hours private transport satisfaction .27 .0 .27

Observed variables

Car owners .00 .03 .03 Health .10 .16 .26

Neighborhood safety satisfaction .00 .19 .19

Satisfaction with neighborhood infrastructure for .00 .16 .16 walking/biking

Age .00 .07 .07

Single .00 .01 .01

Male .00 -.005 -.005

Household income .00 -.04 -.04

Households with kids .00 .05 .05

Non-traveler .42 .00 .42

Overall neighborhood satisfaction .39 .00 .39

108

Transport well-being

Table 6.5 Relationships among the latent and observed variables

Endogenous Observed Standardized Unstandardized Squared

Latent variables Coefficients coefficient t p multiple variables correlations

Travel cost .97 1.00 .93 Travel time .99 1.10 48.62 *** .97

Waiting time .99 1.20 45.83 *** .97 Public Crowdedness .97 .99 39.81 *** .94 transport Distance to .99 1.3 47.66 *** .98 satisfaction station/stop

Safety 1.00 1.19 50.51 *** .99

Inside means .99 1.12 44.23 *** .98 quality

Parking .99 1.10 36.50 *** .97 availability

Traffic .96 .91 31.02 *** .92 Private congestion

transport Travel time .95 .97 33.85 *** .90

satisfaction Safety .96 1.05 40.41 *** .92

Parking cost .86 .84 21.69 *** .76

Travel cost .95 1.00 .89

Travel time to .22 .32 2.54 *** .12 work

Travel time to .58 .91 6.15 *** .35 service places

Travel time to .22 .30 2.91 *** .29 Accessibility kids school/day satisfaction care

Travel time to .67 .74 6.50 *** .46 family/friends

Travel time to .70 1.0 .49 recreational places

109

Transport well-being bigger than 1.96 for p<.05 and bigger than 2.33 for p<.01. The other main results are listed below:

 All the trip attributes (e.g. travel time, travel safety, etc.) for both public and private means significantly explain the associated latent variables, and thereby link to travel satisfaction.

 Satisfaction with “Parking availability” is the most dominant attribute among the predictors of “Peak hours private transport satisfaction”.

 Satisfaction with “Safety” is the most dominant attribute among the predictors of “Peak hours public transport satisfaction”.

 Satisfaction with “Travel time to recreational places” and “Travel time to family and friends” are the most dominant attributes of “Accessibility satisfaction”.

6.6 Interpretation

Our results show that in general all the variables in our model are significantly correlated with travel satisfaction. Below these correlations will be discussed in more details.

6.6.1 Effects of travel characteristics

Our model results suggest that rush hour commuting and mode of transport significantly contribute to travel satisfaction. Non-traveler in peak hours are in average happier with their travel than car commuters and public means commuters. The high amount of people who does not commute in peak hours is also related to the fact that working from home is popular in the Netherlands. The lower level of travel satisfaction among public transport commuters in compared with car commuters also admitted by several other researchers (e.g. St-Louis et al., 2014; De Vos et al., 2016). For instance, Susilo and Cats (2014) who examined the key determinant of travel satisfaction for multi-stages trip found that commuters have low satisfaction levels with public transport trip stages while they are happy with the trip stages by other means of transport. Regarding the trip attributes, car commute satisfaction in peak hours is strongly explained by satisfaction with parking availability and traffic congestion, while satisfaction with parking cost is the least important attribute. This might be due to the availability of free parking for most of working people, who are a big proportion of car commuters during peak hours. Public transport satisfaction in peak hours is strongly

110

Transport well-being

the SEM model with standardized coefficients for transport satisfaction transport coefficients for standardized model theSEM with

Results of Results

6 Figure

111

Transport well-being explained by seven attributes of which satisfaction with safety has the highest level of importance while satisfaction with trip cost has the lowest importance.

6.6.2 Effects of socio-demographic variables

Regarding the impact of socio-demographic variables, similar to many other studies, we found many significant associations. Age of commuters positively impacts travel satisfaction ratings which is similar to findings of many other researchers (e.g. Ory & Mokhtarian, 2005; Gao et al., 2016). However older people are less likely to be happy with their travel time to work and car commuting in peak hours. Furthermore, people with higher self-reported health are more satisfied with their travel, which is also admitted by Ye and Titheridge (2016) who examined health as one of the socio- demographic variables, and found it as an influential attribute on travel satisfaction. Beside the direct effect, our results show that health does affect travel satisfaction indirectly through accessibility satisfaction, neighborhood satisfaction, safety feeling and satisfaction with physical activity, which in turn influence travel satisfaction.

In terms of gender, the relationships are more complicated as female commuters are slightly happier with their overall transport, and they reported higher satisfaction with accessibility feeling while male commuters have higher satisfaction levels with safety assessment in public transport commute (e.g. Yavuz & Welch, 2010), and also are happier with car commuting in peak hours. But eventually our results show that female respondents are happier with overall travel satisfaction than male respondents, which could be due to the fact that female respondents travel less in peak hours. Gao et al., (2017) who conducted a research on the effects of mood and personality on satisfaction ratings with daily trips also found that male travelers are more likely to feel anger and be critical, and thereby have lower level of travel satisfaction.

Single people have higher satisfaction level with the accessibility and car commute in peak hours, and thereby have higher level of travel satisfaction. Car ownership has been shown to have direct influence on mode choice and positively impacts travel satisfaction. It also influences neighborhood satisfaction, satisfaction with travel time to family/friends and recreational centers. Our results also indicate that presence of kids at households leads to higher level of travel satisfaction. In general, our results have shown that female, older people, single people, car owners and household with kids are more likely to be happy with their overall travel than others. Further, as we already discussed the effects of travel mode choice and rush hour

112

Transport well-being commuting on travel satisfaction, it is good to note that socio-demographic characteristics of respondents are associated with commuting choices which in turn influence travel satisfaction. For instance, younger people and males are more likely to travel during peak hours. These associations, to some extent explain the indirect effects of socio-demographic variables on travel satisfaction.

However, although socio-demographic variables may significantly influence travel satisfaction as their effects reflects differences in underlying needs, desires and space-time constraints, but psychological factors associated with certain socio- demographic characteristics may play a more critical role. For instance, older people are generally more relaxed and thereby have higher level of travel satisfaction. In this regard, personality has a more direct and clear link with travel satisfaction than socio- demographic variables (Gao et al., 2017).

6.6.3 Effects of built environment

It has been shown that most of built environment characteristics are potentially associated with travel mode choice and trip attributes (e.g. Cao & Ettema, 2014; De Vos et al., 2016; Ye & Titheridge, 2017), which in turn influence travel satisfaction. In our model, beside the direct effects of neighborhood satisfaction and accessibility on travel satisfaction, built environment characteristics indirectly influence travel satisfaction through satisfaction with car commute travel time and parking cost. Accessibility explains the extent to which people are happy with travel time to important places such as work, recreational places, etc. And neighborhood characteristic such as safety and infrastructure for physical activity influence satisfaction with neighborhood, and in turn affect travel satisfaction. Car ownership is significantly correlated with neighborhood satisfaction and accessibility variables, particularly accessibility to family/friends and recreational centers, while it has no significant relationship with travel to work, kids school and service places. We can conclude that the decision of car ownership is related to residential locations and residential choices are likely to be based on distance to work places, etc. Therefore, satisfaction with residential location and accessibility meditate the effect of car ownership on travel satisfaction.

113

Transport well-being

6.7 Conclusions and discussion

Despite extensive research on travel behavior, the study of travel satisfaction is relatively new. These studies mostly focus on the assessments of travel characteristics. In this research project, satisfaction with the built environment and access to important places in daily life has also been included. Our empirical results provide evidence that the proposed conceptual model is acceptable. Nevertheless, data limitations and small sample size for some modes of transport is a big challenge in interpreting the results as travel satisfaction is just one domain of our research.

Although context differences may lead to differences in travel characteristics (travel mode choice, level of service, etc.) and built environment characteristics among different places, however, several findings of this study are in line with previous studies conducted in other countries. These findings include: (1) travel mode choice is significantly associated with travel satisfaction, (2) rush hours trips have a significant effect on travel satisfaction, and (3) healthier commuters are happier with built environment characteristics, and health of commuters directly and indirectly impacts travel satisfaction. However, to the best of our knowledge, there are few findings that are unique to this study. These findings include: (1) satisfaction with the built environment affects travel satisfaction directly and indirectly through the path of travel characteristics (e.g. travel time), (2) a high level of rush hours avoidance, which directly influences travel satisfaction is related to work flexibility in the Netherlands. Working from home is a popular trend, and thereby, part of non-travelers during peak hours concern employed people. This indicates that work flexibility directly influence travel satisfaction among Dutch commuters.

114

Work well-being

7.1 Introduction

Work is a dominant sphere of life. It is the main activity that individuals across the globe perform to generate income and pay for life’s necessities and living standards. It is also closely associated with non-material dimensions of life such as social life and integrating individuals into society (Fryers, 2006). At a higher level, work is related to one’s goals, desires and achievements and from this perspective, it is more than just a regular activity, but can be a source to shape identity and personal growth (Hulin, 2002). Thus, it is a rewarding activity and pleasure of doing a desirable work is a significant part of life satisfaction. In that sense, many empirical studies have shown that work satisfaction is strongly associated with overall life well-being (e.g. Sirgy., 2012; Lee et al., 2015; Senasu & Singhapakdi, 2017). Along the same line, results from many studies have demonstrated that the lack of work or being unhappy with one’s

115

Work well-being career can have many negative impacts on life and deteriorate well-being (e.g. Gerdtham & Johannesson, 2001; Abdallah et al., 2012).

Despite the considerable role that job satisfaction plays in one’s life, the vast majority of well-being and happiness studies have typically focused on income and financial aspects of the job. These studies ignore important features of job satisfaction such as social and psychological aspects, which contribute to fulfillments of higher level of needs (Diener & Seligman, 2004). On the other hand, most of the literature on work satisfaction is a part of economic and organizational studies with the ultimate target of optimizing work performance and effectiveness in business organizations rather than enhancing employees’ well-being. This little emphasis on work satisfaction as a multifaceted concept implies a gap in the literature and the need for more detailed and comprehensive research.

In formulating the conceptual model of this PhD research (see Chapter 2), work as a main human activity plays a key role in quality of life, and thus, this chapter extends the knowledge derived from prior research by investigating the role of different attributes on work satisfaction. More specifically, in this chapter we will address the following research questions:

I. What are the main predictors of work satisfaction?

II. To what extent do these indicators contribute to predicting work satisfaction?

III. What are the relationships among these indicators and to what extent do they affect each other?

7.2 Literature review

7.2.1 Activity called job (terminology and definitions)

There are numerous terms to explain the regular activity that generates income. Among those, job, career, profession, work and employment are the most common terms that used interchangeably in the organizational literature. Duran (2015) argued that “job” refers to work as a paid position and thereby highlights the earning aspects, while “work” covers unpaid activities as well. Also, based on the main definition of work as

116

Work well-being the “Activity involving mental or physical effort done in order to achieve a purpose or result” (Oxford dictionary), it is reasonable to say that “work” reflects more aspects of the notion as it refer to “productivity” and higher demands of people.

However, as there is no consensus among researchers on terminology, work and job satisfaction normally use and refer to similar concepts and various researchers provided their own specific definition. More recently, some authors have used the term “quality of work life (QWL)” to explain the concept with relation to well-being research. For example, Varghese and Jayan (2013) indicated that QWL is the best term to explain the concept of work satisfaction related to subjective well-being. They provided the following definition for QWL: “Quality of Work Life is that part of overall quality of life that is influenced by work. It is more than just job satisfaction or work happiness, but the widest context in which an employee would evaluate their work environment”. We can summarize that the current attempts at defining QWL consider it as a broader ranging concept that covers the overall experience of work, which consists of satisfaction from different work-related components including three main categories of work space, work conditions and work location.

7.2.2 Research background

The emergence of work satisfaction studies can be traced back to early work of Hoppock (1935), where he described job satisfaction as the employees’ subjective reflections to working scenarios (Zhu, 2012). During the twentieth century, investigating the link between employees’ attitudes and efficiency became a fascinating research topic in social psychology and management field. But, the first study that addressed job satisfaction measurement appeared in the work of Churchill et al. (1974), who explained the job satisfaction concept via features of job and features of job-related environment. However, during the years, the concept evolved gradually from single affective perspective to multiple perspectives containing both affective and cognitive dimensions. The current focus of most job satisfaction studies are on turnover intention (e.g. Saeed et al., 2014; Sukriket, 2014), and the importance of work-life balance (e.g. Haar et al., 2014; Sirgy & Lee, 2016; Mas-Machuca et al., 2016) in different occupational contexts. In this regards, many studies found that the lack of work-life balance tends to increase the turnover intentions among employees (e.g. Suifan et al., 2016).

117

Work well-being

7.2.3 Theoretical concepts

Theories that explain work satisfaction have strong roots in theories on human motivation. One of the main theories is Maslow’s needs hierarchy theory. Based on this theory, the concept of Quality of Life (QOL) has been defined as “the degree to which the experience of an individual’s life satisfies that individual’s wants and needs” (Rice, 1984). In this definition, one major domain of life experience is considered as the experience of work. Having a satisfying work life can respond to basic needs as well as higher level of human development needs such as “self-actualization,” which means the full realization of one’s potential, which is posited at the top of the Maslow’s hierarchy of needs. In other words, work can contribute to all layers of human needs by providing salary, secure work environment, friendly environment, job title, and challenging job (Sirgy, 2012). Thus, some authors apply Maslow’s theory of hierarchical needs to explain work life satisfaction. Other theories such as Locke’s Range of Affect Theory (1968), Herzberg’s two-factor theory (Alshmemri et al., 2017), and Job characteristics Model (JCM) (Hackman et al., 2005) have also been used in many empirical studies.

Range of Affect Theory basically determines satisfaction by the difference between what an employee wants from job and what employment offers in return. It also contends employees’ prioritization of one job facet over others. Herzberg’s two- factor theory or motivation-hygiene theory is a two dimensional theory that explains the difference between two groups of factors: one group are motivation factors (such as advancement, recognition etc.) that can enhance job satisfaction, and the other group are hygiene factors (related to the workplace and environment) with the ability of reducing job dissatisfaction (Alshmemri et al., 2017). Moreover, job characteristics model (JCM) listed five core job characteristics that can influence critical psychological states. These psychological states (experienced meaningfulness, experienced responsibility, and knowledge of the results activities), in turn, impact motivation, job performance and job satisfaction. However, all of these theories have drawn criticisms and none can fully explain work satisfaction, since some focus on the impact of the surrounding environment (the motivational theories), while others focus on personality (such as JCM). Furnham et al. (2009) proved that personality and demographic variables of individuals together with the surrounding environment can account for a significant portion of work satisfaction.

118

Work well-being

7.3 Factors determining work satisfaction

There are numerous factors that highly impact work satisfaction. Based on an extensive available literature, socio-demographic factors including several work-related aspects influence judgments regarding employment (Table 7.1). Individual characteristics such as age, gender, marriage status, education, and job characteristics such as satisfaction with pay level, workload and work quality including the physical characteristics of workplace are the most discussed variables (Steijn, 2004). Here, we explore these variables separately, and in more details.

7.3.1 Individuals’ socio-demographic factors

Age is one of the mentioned factors that might influence work satisfaction. As age and career stage are related, it is relevant to say that age tends to be positively correlated with job satisfaction. Moreover, some researchers (e.g. Zacher & Griffin, 2015) revealed that age influences career adoptability and thereby increases job satisfaction. However, in contrast to this result, some studies found no significant relation between job satisfaction and age (e.g. Chaudhuri et al., 2015) whilst others found job satisfaction is U-shaped in relation to age (e.g. Clark et al., 1996; Gazioglu & Tansel, 2006).

Regarding gender differences, a number of studies show that females are more satisfied with their jobs in general than males (e.g. Clark, 2005; Selvarajan et al., 2015). However, the results are mixed, particularly when it comes to various aspects of job evaluation (e.g. Gazioglu & Tansel, 2006). For instance, some studies found that women are more likely to be satisfied with their salaries (Crossman & Abou-Zaki, 2003; Hauret & Williams, 2017), but less satisfied with the opportunities for work advancement (Carvajal et al., 2018). There are also many studies that found no significant effect of gender differences on job satisfaction when all job conditions are similar (e.g. Spencer et al., 2016).

The relationship between marital status and job satisfaction show that career and family life are strongly tied to each other. In fact, both family life happiness and job satisfaction impact each other. Similar to other socio-demographic factors, the findings regarding the impact of marital status on job satisfaction are in conflict. For instance, Gazioglu and Tansel (2006) found married employees are less satisfied than single employees. In contrast, Hagedorn (2000) revealed that married employees reported higher level of work satisfaction. Moreover, other socio-demographic factors such as

119

Work well-being

Table 7.1 Major attributes in the selected recent literature

Study Indicators

 Salary  Stress  Workload

 Opportunity for advancement  Job security Carvajal et al. (2018)  Fairness  Job atmosphere  Relations with co-workers  Supervision  Scheduling flexibility  Autonomy  Importance of job  Work environment,  Organization culture and climate,  Relation and co-operation, Nanjundeswaraswamy  Training and development, & Swamy (2013)  Compensation and Rewards,  Facilities,  Job security,  Autonomy of work,  Adequacy of resources.  Adequate And Fair Compensation,  Safe And Healthy Working Conditions,  Immediate Opportunity To Use and develop human Hosseini (2010) capacities,  Opportunity For Continued Growth And Security,  Social Integration In The Work Organization,  Constitutionalism In The Work Organization,  Work And Total Life Space,  Social Relevance Of Work Life.  Fair Pay and Autonomy,  Job security,  Fair environment,

 Interesting and satisfying work,  Trust in management, Saraji & Dargahi (2006)  Support from coworkers,  Recognition of efforts,  Career prospect,  Control over work,  Health and safety standards,  Work-life balance,  Amount of work,  Level of stress,  Occupational safety.

120

Work well-being

 Individual characteristics (age, gender, ethnicity, and education level),

 Job characteristics (income, supervisory position, working full time, permanent job, skill utilization and Steijn (2004) sector of work),  Work environment (task autonomy),  Satisfaction with management, pay and workload),

personnel management practices and overall satisfaction.  Poor working environments,  Resident aggression,  Workload, inability to deliver quality of care preferred,  Balance of work and family,  Shift work,  Lack of involvement in decision making,  Professional isolation, Ellis & Pompli (2002)  Lack of recognition,  Poor relationships with supervisor/peers,  Role conflict,  Lack of opportunity to learn new skills.

 Favorable work environment,  Personal growth and autonomy,  Nature of job, Wyatt & Wah (2001)  Stimulating opportunities,  Co-workers.

education, years of work experience and income level of the family have also been investigated in studies of job satisfaction, but no significant associations have been reported.

7.3.2 Employment status (Part time vs full time)

The difference between part-time and full-time workers is the amount of time they spend on work, which can lead to more flexibility on the one hand but a lower salary on the other hand. Research findings regarding the impacts of work status (part-time vs

full-time) on job satisfaction are mixed. Some researchers argue that having a part-time job can lead to a lower job satisfaction due to a lower salary, less stability and promotion opportunities (e.g. Kauhanen & Nätti, 2015; Montero & Rau, 2015). In contrast, some results display higher satisfaction among part-time employees (e.g. Wittmer & Martin, 2011). Karatuna and Basol (2017) acknowledged that there are many

121

Work well-being factors that influence the relation between employment status and job satisfaction, and among those, job characteristics and different career expectations play critical roles. Moreover, Fagan (2001) argued that country-level policies in terms of employment conditions can influence the standards, which are the reference for employment judgments, and thereby work status satisfaction tends to be influenced by the country context.

However, in general it is well documented that part-time employees who actually choose to work part-time have equal or higher job satisfaction compared with of full- time workers because less workload reduce problems with work-life balance, and enables part-time workers to act better in other life domains (Russo & Buonocore, 2012). Thereby, the extent to which employment status influences workers’ satisfaction is dependent on whether the employees want to work part-time (voluntary) or other obligations force them choosing their employment status. Accordingly, research findings show that women who work part-time have higher levels of job satisfaction in comparison with part-time men workers, which is due to a higher preference (because of family responsibilities) for part-time work among women (Booth & van Ours, 2008; Montero & Rau, 2015).

7.3.3 Work conditions

Work conditions category contains the main determinants of job satisfaction and mainly reflects the general nature of work or even the primary reasons of choosing a job. Quality of work and pay level is the two key factors for most people choosing their job. In addition, factors such as social environment, relations with colleagues and workload can only be determined in later stages when individuals commence a job. However, these indicators might even play a higher role in one’s assessment of job satisfaction.

7.3.3.1 Pay level

Pay level has a motivational role for any job and it is a primary reason for most individuals to devote themselves to work. Because of this significant role, many researchers since the 1960’s investigated the role of pay level satisfaction on job satisfaction (e.g. Locke, 1968; Judge et al., 2010). Despite many reasons that theoretically support the strong impact of pay level satisfaction on job satisfaction, there are conflicts among the findings in empirical research. Some researchers found a significant positive link between payroll satisfaction and job satisfaction (e.g. Darma & Supriyanto, 2017), whilst others suggest that having a well-paying job is only marginally 122

Work well-being related to job satisfaction. However, the conflicting part is the actual size of the relationship of pay level satisfaction to job satisfaction rather than the relation itself.

Cognitive evaluation of pay level has a complicated mechanism according to one’s expectations. On the one hand, there should be a balance between job characteristics such as working time, job difficulty, etc. and the pay level to make individuals happy about their salary. On the other hand, there are other psychological factors that influence pay level satisfaction, such as distributive justice perception. Levy and Norris-Watts (2004) argued that employees have a certain perception regarding the fairness of their salary. Following that, Judge et al. (2010), indicated that employees compare their earnings with other co-workers, so they are likely to feel satisfied of their salary when they get paid more in comparison with others. These findings highlight the fact that earning satisfaction is not only related to the amount of pay, but the judgments can be relative to other reference points and external factors.

7.3.3.2 Quality of work

Perceived quality of job is a significant facet of employees’ career and reflects how employees feel about job tasks. It deals with the nature of work, its features and the personal characteristics of employees. According to Clark (2015), quality of work is under the influence of a series of factors such as autonomy, opportunities for advancement and interest in the job. It is said that, when the education and experience of employees are matched with the job level and position, the performance is more effective. Thereby, during the recent decades, many organizations tend to measure and enhance work quality satisfaction of their employees. In this context, scholarly findings (e.g. Clark, 2005) advocate that quality of work is more important than wages. On the one hand, job is a goal-directed activity, depending on whether it provides employees with sense of progress and efficiency; it can be a source for happiness. Available evidence demonstrates that less favorable job quality is often associated with lower work related well-being (Van Aerden et al., 2015).

7.3.3.3 Workload and job stress

Existing research also points to workload satisfaction as a significant predictor of job satisfaction. Workload in general can be defined as “the amount of work that should be done in a certain period of time and with a certain quality” (Sahin & Sahingoz, 2013). Whilst the amount of work a person can handle differs among people based on physiological and personal characteristics, researchers agreed on the association of 123

Work well-being longer working hours with higher levels of job dissatisfaction and risk of turnover (Brannon et al., 2007; Qureshi et al., 2013). In fact, feeling overworked can cause higher health risks and job stress (e.g. Haggan, 2017). In addition to that, heavy workload can jeopardize the work-life balance and family life, as it is more difficult for employees to devote enough time for their personal interests (Pocock, 2005). Consistent with this, workload is not just a predictor of job satisfaction but in a wider context, it has strong correlations with personal well-being (Bowling et al., 2010).

7.3.3.4 Social environment (people at work)

The relationship with colleagues is an important aspect of work satisfaction. Adams & Bond (2000) and Jain & Kaur (2014) indicated that there is a strong link between interpersonal relationships at work and job satisfaction. Furthermore, in the quest for greater productivity, many organizational researchers (e.g. Harris & Kacmar, 2006) argued that good social relations at work can reduce job-related stress and can allow employees to deal with their jobs more efficiently. There is a strand of literature in job psychology that refers to social relations as a function of exchanging information and knowledge sharing. The concept of “knowledge sharing” is basically defined as a helping attitude of employees that increases the intellectual capabilities and can encourage a positive feeling regarding the job (e.g. Jiang & Hu, 2016). At a higher level, getting support and help from collogues at work can facilitate the fulfillment of human need for relatedness. Erdogan et al. (2012) stated that there is a link between the social environment at work place and the overall life well-being. Wasko and Faraj (2005) also argued that most people spend a huge portion of their life time at their work place and therefore, the impacts of work-related social life is beyond job satisfaction and can play a significant role in overall life well-being.

7.3.4 Workspace

Unlike work location, workspace satisfaction has received extensive attention from organizational and occupational scientists as well as environmental psychologists. Especially in recent years, the emergence of economic concepts such as open-plan office has brought new arguments to the field and thereby, a large body of research has investigated the contribution of the physical environment to the performance of the employees and their job satisfaction (e.g. Fornara et al., 2006; Aries et al., 2010; Veitch et al., 2011; Lottrup et al., 2015). It is well documented that functionally uncomfortable and poorly designed workspaces decrease both productivity (e.g. Dole & Schroeder, 124

Work well-being

2001) and job satisfaction (e.g. Djukic,et al., 2014). In general, workspace satisfaction is the extent to which an employee feels the space-related needs for performing the job tasks are fulfilled. In fact, workspace characteristics can motivate employees to handle their job better. Furthermore, Vischer (2007) argued that a good workplace should achieve environmental comfort at three level of physical, functional and psychological. Accordingly, other researchers (e.g. Fornara et al., 2006; Rioux, 2017) noted that the perceived evaluation of workspace is an interaction of physical and functional characteristics of the environment with psychological and affective attributes. Moreover, how employees evaluate their workspace is very much dependent on their personal characteristics, culture and the comparison between their current workplace and ideal workspace (Marcouyeux & Fleury-Bahi, 2011). In brief, workspace has some important components, such as size, noise, privacy including the availability of necessary facilities (to adjust light and temperature) and tools (proper desk, chair, computer, etc.), which have been considered frequently in empirical research.

Size of workspace refers to the necessary space every employee is entitled to have to perform tasks and move easily. Close proximity to other co-workers produces occupational hazards and thus should be avoided. Workplace density and size are thereby intertwined. Space allocation standards and regulations clearly determine the minimum space that employees need for work (especially for office work), and the standards of the total net area that is shared by staff. Apart from the necessary space that every employee needs, the acoustic environment and low noise levels are other factors that certainly contribute to the employees’ performance and workspace satisfaction (e.g. Dawal & Taha, 2006). Noise, in general, is one of the main recognized factors of the environmental stress, which is associated with a range of negative health effects. Loudness and intensive sound decrease concentration and contribute to distraction, which may highly reduce motivation among employees (Rioux, 2017). Respecting current standards and recommendations (e.g. ISO standard), the level of noise exposure at workplaces should be limited, and employers should take preventive actions (such as using sound proofed walls or other equipment) to control the noise level of the workspaces.

Privacy is another significant factor associated with employees’ workplace satisfaction. Privacy at the workspace has become a very controversial topic as the open office plan has become popular and many believe that the absence of walls and partitions may threat feelings of privacy and workspace satisfaction (e.g. de Croon et

125

Work well-being al., 2005; Danielsson & Bodin, 2009). Although boundaries and content of what is private at workplace are not well defined but the concept of privacy at workplace usually refers to the ability of performing tasks without unwanted distraction by others. Moreover, Birnholtz et al. (2007) noted that privacy could be seen from two different perspectives: optimized confidentiality and solitude. Whilst the former refers to control over accessibility of others to employees’ information, the latter refers to the limitation of distraction (visual, acoustic and territorial) by others.

Existing research also points to workplace facilities, tools and furniture as important factors influencing employees’ performance and satisfaction (e.g. Vischer & Fischer, 2005; Nguyen et al., 2015). In addition, many researchers (e.g. Lee & Brand, 2010) found that ambient workplace conditions, such as temperature and lighting can have significant impacts on the employees’ attitudes at work. Furniture should be flexible, adjustable and suitable for performing tasks. Also, furniture should properly respond to ergonomic needs of employees, which in the long term affect their personal health (Vischer, 2007). Accordingly, ambient conditions such as lighting, ventilation and the thermal system should have the possibility of adjustment based on employees’ requirements and comfort level.

7.3.5 Work location

Just like the house location, the working place location has important implications for employees’ well-being. In fact, residential choice and workplace location are intimately interconnected, as job location decisions are usually based on residential location and vice versa (Clark et al., 1999; Hincks & Wong, 2010; Head & Lloyd-Ellis, 2012). Work location dissatisfaction can cause housing mobility and otherwise decrease overall job satisfaction, which can lead to employees’ turnover. Living close to the job place or having good accessibility can benefit employees by saving time, energy and cost of commuting. Moreover, in recent decades, behavioral scientists have addressed the impacts of work commutes on daily life experience. It has been found that having good access to the work place can significantly affect daily life experiences. Kahneman et al. (2004), in their well-known research, using the day reconstruction method, found that commuting to the work episode of daily life is associated with the most negative feelings among employees. Despite the well-established link between the work commute and well-being in travel behavior research and urban science (e.g. Olsson et al., 2013), organizational research displays a certain lack of establishing the link

126

Work well-being between work commute and work satisfaction. Only few studies have investigated work commute as a key generator of stress and a reason of turnover (high level of job dissatisfaction) among employees (Amponsah-Tawiah, 2013). Also, from a travel behavior perspective, Gao et al. (2017) also found that travel satisfaction was not significantly related to life satisfaction. However, there is much room left for further research in this subject.

7.3.5.1 Travel time

Commute time to work constitutes a large part of daily routine activities and significantly impacts employees’ well-being. Several studies have provided empirical evidence that long travel times make employees feel exhausted and unhappy and contribute to lower productivity (e.g. Olsson et al., 2013; Handy & Thigpen, 2018). In addition, commute stress in long trips can affect physical and mental health, which consequently diminishes well-being and work life satisfaction (Novaco & Gonzalez, 2009). To reduce commuting time, telecommuting became a popular trend in recent years. Telecommuting enables employees to work from an alternative work place (usually house) and proactively manage their schedule (Allen et al., 2015). It is argued by many researchers that telecommuting has a positive influence on job satisfaction by minimizing work-life conflict and job stress (e.g. Fonner & Roloff, 2010).

7.3.5.2 Other work location attributes

It is accepted that travel time to work is the most commonly mentioned attribute of perception regarding job location (with relation to house location), and has the largest influence on job satisfaction. However, it is not proper to neglect the role of other characteristics of the work location on perceived satisfaction of the work location. These characteristics comprise safety of the surrounding area, access to alternative transport modes and availability of facilities such as shops, restaurants, parks etc. These factors help increasing job attractiveness by providing conditions to combine activities with work such as going out for fresh air to reenergize the workday, small grocery shopping or performing leisure activities with colleagues (such as a short walk). Despite the fact that these factors can play an important role in facilitating the work-life balance and job satisfaction, it is hard to find any research that combines these factors to predict job satisfaction. This might be due to the great emphasize on salary, workload, etc. or the primary assumption that employees should spend most of their time inside the office rather than outside and thereby do not care about outside qualities. 127

Work well-being

7.4 Analysis and results: Multiple regression analysis

In order to explore the implicit effect of various work dimensions on the employees’ overall assessment of their work, first, a multiple regression analysis was conducted, and then, a comprehensive analysis using path analysis is being applied. Multiple regression analysis is done using overall work satisfaction as the dependent variable and 17 factors as independent factors (including socio-demographic variables).

We assumed that the scales used to measure satisfaction have interval properties (same degree of difference between items) based on many studies that utilized such scales (Molin & Timmermans, 2003). Since one of our regression assumptions is that the residuals (prediction errors) are normally distributed, checking the histogram of residuals allows us to check the extent to which the residuals are normally distributed. The well-behaved residuals can lead to correct p values, which are very important for any linear regression analysis. Our histogram (Figure 7.1) shows a fairly normal distribution. Thus, based on these results, the normality of residuals assumption is met. Furthermore, the Normal P-P Plot of the regression residuals (Figure 7.2) is a good tool to assess the assumption of normally distributed residuals. If data points (little circles) follow the diagonal line, then the residuals are normally distributed. Our P-P plot shows minor deviation and thus the normality of residuals assumption is confirmed. The results about R-squared (R2) and adjusted R-squared which are shown in Table 7.2 reveals that the indicators accounted for almost 70 percent of variance of overall work satisfaction.

7.4.1 Interpretation

Regression coefficients in this analysis represent the magnitude of different factors that influence employees’ overall work satisfaction, and are presented in Table 7.3. For example, the factor that shows the strongest relationship with overall work satisfaction is satisfaction with quality of work (.34) following by privacy satisfaction (.26). Although multiple regression analysis provides some insights into the impact of socio- demographic characteristics and all of the mentioned work-related characteristics on overall work satisfaction, this method is unable to capture the indirect and complex relationships among work location, work conditions and workspace satisfaction factors. Thereby, further analysis is necessary to understand the mechanism of the effects among these main categories as well as their internal dependencies, which will be explained in the next part.

128

Work well-being

Figure 7.1 The Histogram of the overall work satisfaction

Figure 7.2 Normal Probability Plot (P-P) of the regression standardized residual dependent variable for overall work satisfaction

129

Work well-being

Table 7.2 Goodness-of-fit of the regression model for overall work satisfaction

R R2 Adjusted R2 Std. Error of the Estimate

.859 .737 .697 .526

Table 7.3 The estimated parameters of the work satisfaction using multiple regression analysis

Unstandardized Standardized

Coefficients Coefficients

Model Parameters B Std. Error Beta t Sig.

(Constant) .999 .511 1.955 .053 -.070 Socio- Age -.005 .004 -1.317 .191 -.087 demographic Single person -.176 .113 -1.563 .121 -.023 indicators Part-time employee -.046 .109 -.418 .677 -.040 Household income -.025 .033 -.748 .456

Quality of work satisfaction .266 .055 .340 4.821 .000

Work conditions social environment .091 .045 .131 2.012 .047 satisfaction satisfaction

Salary satisfaction .028 .038 .044 .744 .459

Workload satisfaction .088 .039 .135 2.291 .024

Access to work space .009 .061 .011 .144 .886 Workspace satisfaction

satisfaction Privacy satisfaction .137 .042 .263 3.227 .002

Size satisfaction .049 .050 .070 .982 .328

Noise level satisfaction .035 .036 .068 .981 .329

furniture/tools satisfaction .065 .040 .100 1.604 .112

Availability of shops satisfaction .042 .031 .088 1.384 .169

Work location Travel time satisfaction .055 .026 .125 2.104 .038

satisfaction Safety of area satisfaction .005 .045 .007 .116 .908

Access to public transport .050 .027 .121 1.885 .062 satisfaction

130

Work well-being

Figure 7.3 The main determinants of the work satisfaction used in this study

7.5 Analysis and results: Path analysis

The key objectives of this chapter are to evaluate and understand the interrelationships among study variables affecting work satisfaction. A path analysis is deemed to be an adequate solution for the achievement of these objectives, thus is used for a detail analysis on work satisfaction. By this method, it is possible to conduct simultaneous assessments of the associations among multiple dependent and independent attributes and thus increase the statistical efficiency (Ringle & Sarstedt, 2014).

It is good to mention that some of the previous job studies performed factor analysis prior to path analysis because it is required to reduce extensive number of variables (that are used to quantify the immeasurable psychological concepts). However, in our study based on the extensive literature review, three major categories of job satisfaction have defined and within each category certain measurable variables have been extracted. Therefore, there is no need to conduct factor analysis prior to path analysis.

131

Work well-being

7.5.1 Conceptualization

To develop our conceptual model, first the main determinants of work satisfaction are categorized in three different groups (Figure 7.3). The first category is work conditions satisfaction, which consists of four main elements of workload satisfaction, work quality satisfaction, salary satisfaction and social environment satisfaction. The second group is the workspace category with five determinates of accessibility of workspace, size, privacy, noise and furniture and tools satisfaction. And finally the third category, which is the work location satisfaction, combines satisfaction of travel time, access to public transport, safety of area and availability of shops and other facilities in the area.

Based on the insights from the work satisfaction literature in organizational behavior research, as well as environmental psychology and well-being research, and the relevance of these studies to investigate the link between different work-related attributes, socio-demographic factors and the overall subjective assessment of work, the conceptual model elaborated (Figure 7.4). The dependent variable of overall work satisfaction thereby, will be studied by exploring the relationships among different attributes and different categories including the effects of independent factors of socio- demographic characteristics of individuals. Figure 7.4 presents the conceptual model of this study.

7.5.2 Path analysis model for work satisfaction

Similar to other chapters, to estimate the conceptual model, software package AMOS 22 is employed. Path analysis with Maximum likelihood approach of parameter estimation is performed to assess: the direct and indirect correlations among factors, examine P- values of the paths, and the standardized regression weights. Maximum likelihood is a robust method, which assumes multivariate normality. In the first step, all of the socio- demographic factors were included in the model to see whether they have any direct or indirect relationship with other endogenous variables. However, after removing the insignificant relationships from the path model, just four variables of socio-demographic group (age, marital status, income and employment status) are remained in the model. This is in line with the findings from literature that mentioning most of the socio- demographic factors have little or no impact on work satisfaction of individuals.

Since the aim of path analysis is to figure out how exogenous variables work together and which paths are important in affecting the endogenous variables, for big models checking the modification indices is one way to improve the initial hypothesized

132

Work well-being

Figure 7.4 The conceptual model model fit. This means that by adding these suggested paths or covariance among the variables or the error terms, several models can be developed, tested and compared to each other. Here, the challenging part is selecting the final model, while all of the options display a good fit despite their distinctions. Looking for an answer among empirical studies did not provide a clear answer as most of research just presents the final and optimal model without explaining the selection process or their criteria. However, some authors mentioned that the only criteria that can explain adding a path, variances or covariance is whether they have theoretical bases or not (Moss, 2009). In line with this, just meaningful paths have added to our model and others are ignored, so the final model is consistent with the available theories.

The results of the final path model (Table 7.4) demonstrate that the model excellently fit the data. Chi Square divided by the model degrees of freedom (/DF or CMIN/DF), which is an essential measure, and should to be less than 2 and preferably close to 1 (Ullman, 2006), equals 1.29 for our model. In addition, the Root Mean Square Error of Approximation (RMSEA) should be less than .08 and ideally less than .05 to

133

Work well-being

Table 7.4 Goodness of the fit for the path analysis model

Fit index Estimate

Chi-square (CMIN) 187

Degrees of freedom (DF) 144

Chi-square / degrees of freedom 1.30

Root mean square error of approximation (RMSEA) .048 CFI .96

IFI .97

TLI .95

represent a great fit (Washington et al., 2010), and in our model RMSEA value is .048. Other indicators of a good fit as mentioned by many researchers (e.g. Moss, 2016) are indices that compare target model with the null model such as: Comparative fit index (CFI), Incremental fit index (IFI), Tucker Lewis index (TLI), Normed fit index (NFI) and Goodness of Fit Index (GFI). The model tends to generate an acceptable fit, if these indexes exceed.90. However, as many researchers mentioned NFI and GFI indices are explicitly sensitive to small sample size below 200, while RMSEA and CFI seem to be less sensitive to sample size (Hooper et al., 2008). Since our sample size in this chapter is 130, to reduce overestimation we considered indexes that are less sensitive to sample size. Moreover, R squared (R2) value of the endogenous variables (called squared multiple correlation value in AMOS), specifically for the ultimate endogenous variables is an additional important criteria for confirming a good model. Our proposed theoretical model satisfactorily accounts for 73 percent of the variance in workers’ overall job satisfaction (R2= .73). Also, R2 of the other dependent endogenous variables (the overall satisfaction of each category) reflects that the determinants for each category are good predictors, as for work conditions satisfaction R2= .72, for workplace satisfaction R2= .68 and for work location satisfaction R2= .71.

Furthermore, statistically significant relationships are found among the following model variables: (1) Satisfaction with the overall work conditions significantly influences the overall work satisfaction (.49), (2) work location satisfaction affects the overall work satisfaction positively (.27), (3) Workspace satisfaction influences work conditions

134

Work well-being satisfaction (.22), work location satisfaction (.19), and the overall work satisfaction (.12).

In the following the other main results will be summarized:  In the work location category, travel time satisfaction is the most influential variable (.41) on work location satisfaction, followed by satisfaction with access to public means of transport (.30).  In predicting work conditions satisfaction, work quality satisfaction (.50) is the dominant variable. Considering the workspace category, privacy satisfaction (.34) and furniture and tools satisfaction (.31) are the most critical factors in predicting workspace satisfaction.  Exogenous variables of socio-demographic group influence variables of the work conditions category and work location category, but not workspace variables.  Also, age is the only variable that directly impacts the satisfaction with work location. However, age and marital status show very small and indirect effects on the overall work satisfaction (.04 and .05, respectively) and only household income has a significant total effect (.10) through mediating variables (i.e. salary satisfaction and work quality).  Obviously, having a part time job status means less work, and thus it has positive influence (.16) on the perception of workload. It has also small effect on overall work satisfaction (.02).  Among the variables situated in one individual category, only social environment satisfaction and work quality satisfaction have strong correlations with each other, which mean that social environment satisfaction has both direct and indirect effects on work conditions satisfaction. It also has a direct effect (.14) on the overall work satisfaction. Privacy satisfaction is another variable with both direct (.22) and indirect effects on the overall satisfaction of work (Figure 7.5).

7.5.3 Interpretation and discussion

Generally speaking, our results are consistent with previous studies emphasizing on the impacts of work conditions components such as work quality and social environment

135

Work well-being

(relations with colleagues) on feeling satisfied with work (e.g. Clark, 2005; Judge et al., 2010; Jain & Kaur, 2014). In terms of the direct effects on overall work satisfaction, work location satisfaction has the second place after work conditions satisfaction. However, workspace satisfaction has both direct and indirect effects on the overall work satisfaction. The indirect effects are also significant with positive signs, whose magnitude is greater than the direct effect. This is a very interesting result as it highlights the fact that space factors are means to reach a goal (such as improve the work quality) but not the goals, and through the mediating variables they can influence the overall work satisfaction. The opposite side is also relevant, as space related issues become more important to those who experience the lack of proper space. One explanation for these findings can be found in Herzberg’s two-factor theory or motivation-hygiene theory, where he basically mentioned that hygiene variables (i.e. environment and place related variables) influence dissatisfaction more than satisfaction. Here, we discuss the impacts of all the components in different categories.

7.5.3.1 The impact of work conditions factors

Among the work conditions factors, satisfaction with quality of work accounts for 51% of the variance of the work conditions satisfaction, followed by workload satisfaction (.27) and salary satisfaction (.15). This clearly indicates the importance of work quality satisfaction over satisfaction with salary. However, the lower impact of salary satisfaction does not imply that it is not a motivational factor ( e.g. Gerhart & Rynes, 2003; Nyberg et al., 2018). The strong correlation among work quality satisfaction and social environment satisfaction emphasizes on the importance of maintaining good social relations at work. In general, social environment at work can mean different things to different people based on their personality and social structure. It may be described by the amount and quality of relationships with colleagues and managers, teamwork, knowledge sharing etc. and as it is expected it has a huge impact on feeling satisfied with work quality. However, considering the total effects of all the variables, it is a surprising result that satisfaction with social environment have such a substantial effect (.039) on overall work satisfaction. This might be the consequence of the wide definition of social environment for the respondents of the survey and our limitations in asking more detailed questions on this psychological construct.

Apart from the aforementioned results, there is a novel finding about the negative correlation among satisfaction with travel time and satisfaction with social

136

Work well-being

sfaction

Results of the path analysis for work sati analysis for work of the path Results

7 Figure

137

Work well-being

Environment. This negative relationship basically means that the more satisfied people get about their travel time, the less satisfied they tend to be with their social environment. To discover what can be the reason, the original data sets and the comments of the respondents were reconsidered. Many of the respondents especially residents of StrijpS (a smart neighborhood) mentioned that they are self-employed (the Dutch term ZZPERS, zelfstandige zonder personeel means independent with no staff) and combine working with living in studios designed for this type of living. Although telecommuting or working from home is a popular trend in the Netherlands, it can bring isolation to individuals. Therefore, basically for freelancers, work commute is reduced or totally omitted, but social life can be negatively affected in return. This interesting finding is very context dependent, as in many countries telecommuting is less common.

Table 7.5 Standardized path estimates of direct, indirect and total effects

Dependent Paths Direct Indirect Total variables effect effect effects

WCS  OWS .49 .49

WSS  OWS .12 .16 .28 WLS  OWS .27 .27

Privacy satisfaction  OWS .22 .16 .37 Overall SES  OWS .13 .26 .39 work satisfaction (OWS) QWS  OWS .25 .25

Access to workspace .23 .23 satisfaction  OWS

Income  OWS .10 .10

Age  OWS .05 .05

Part-time employee  OWS .02 .02

Single person  OWS .06 .06

Furniture/tools satisfaction .13 .13  OWS

Access to public transport .18 .18 satisfaction  OWS

WLoS  OWS .13 .13

QWS  WCS .50 .50

138

Work well-being

WLoS  WCS .26 .26

SS  WCS .15 .15 Work conditions satisfaction SES  WCS .13 .39 .52 (WCS) WSS  WCS .22 .22

Privacy satisfaction  WCS .20 .20

Access to workspace .26 .26 satisfaction  WCS

Furniture/tools satisfaction .15 .15  WCS

Access to public transport .14 .14 satisfaction  WCS

Income  WCS .20 .20

Size satisfaction  WSS .12 .12 Work space satisfaction Access to workspace .17 .17 (WSS) satisfaction  WSS

Privacy satisfaction  WSS .34 .34

Noise level satisfaction  .15 .15 WSS

Furniture/tools satisfaction .31 .31  WSS

TTS WLS .41 .41

Work location Access to public transport .30 .30 satisfaction satisfaction WLS (WLS) SAS WLS .18 .18

ASS WLS .27 .27

WSS WLS .19 .19

Age WLS .12 .05 .17

Privacy satisfaction QWS .24

Quality of work SES QWS .78 .78 satisfaction (QWS) Access to workspace .33 .33 satisfaction QWS

Income QWS .25 .25

Access to public transport .15 .15 satisfaction QWS

Part-time employee WLoS .15 .15

Work load Income WLoS .14 .14 satisfaction (WLoS) TTS WLoS .17 .17

139

Work well-being

Furniture/tools satisfaction .33 .33 WLoS

Noise level satisfaction .11 .11 WLoS

Income  SS .25 .25 Salary satisfaction Single person  SS .13 .13 (SS) Access to workspace .27 .27 satisfaction  SS

Access to public transport .19 .19 Social environment  SES satisfaction TTS  SES - .16 -.16 (SES) Access to workspace .42 .42 satisfaction  SES

Travel time satisfaction Single person  TTS .22 .22 (TTS) Safety of area Access to work space .38 .38 satisfaction (SAS) satisfaction  SAS

Availability of Age  ASS .18 .18 shops etc. Satisfaction (ASS)

7.5.3.2 The impact of workspace factors

In the workspace category, privacy satisfaction and furniture and tools satisfaction are the two most influential factors, while workspace size is the least important component. This is also expected, as the concepts such as shared office are currently popular (mainly for economic reasons), so employees are adapted to working in smaller places. Nevertheless, with the help of technology, good design and proper furniture they can still own their semi-private spaces. Interestingly, privacy satisfaction contributes to the overall work satisfaction both directly and indirectly through the moderator variable. It impacts the satisfaction with quality of work, which can be described through the relationship between control and independency with performance at work. But in general, it highlights the fact that privacy is an essential prerequisite for employees’ working experience and the perceived value in work satisfaction.

Moreover, the other interesting results related to this category are the impacts of noise level satisfaction and furniture and tools satisfaction on the workload satisfaction variable. This has been theoretically and empirically proved that noise and insufficient 140

Work well-being facilities lead to distraction, failure in time management and handling job demands. Finally, the positive relationship between access to workspace satisfaction and social environment satisfaction indicates that the office location within the building can influence social interactions among employees. Furthermore, access to the workspace is significantly associated with safety feeling, which is also a very interesting finding.

7.5.3.3 The impact of work status

As expected having a part time job can reduce the workload and thereby enhance the perception of workload, while it may impact the pay level and consequently bring lower satisfaction with salary. However, surprisingly having a part time job has no effect on salary satisfaction. This might be due to the fact that people evaluate their salary based on the amount of tasks they handle and the time they allocate to their jobs. Since having a part time job does not negatively influence salary and job quality satisfaction, it is also not expected to influence the overall job satisfaction significantly.

7.5.3.4 The impacts of socio-demographic factors

In terms of the employees’ socio-demographic factors, only household income has a significant indirect effect on overall work satisfaction through mediating variables (i.e. salary satisfaction and work quality). The impact of income on salary satisfaction is clear and well documented in the literature (e.g. Judge et al., 2010). In terms of the relation between income and satisfaction with quality of work, we can say that having a higher household income gives the opportunity to work in a favorable job (in case of dual earners households for instance). If the household income is equal to the individual salary (in case of single earner families), also the relation makes sense as pay level has a motivational role on job performance and work quality. Yet, the adverse relation is arguable as enjoying the job, can lead to higher performance and possible pay rise, which consequently increase the household income.

7.5.3.5 Other associations among factors

Some other results can be extracted from our model, although there are beyond the scope of this chapter but it is good to mention them briefly:  Age and having a higher income household have a positive relationship, means that in average people reach to the better financial position by getting aged.

141

Work well-being

 Being single shows a negative relationship with part time job variable. This indicates single people are less likely to have a part time job. Also as assumed, household with higher income may not work in part time jobs.  Single people are more satisfied with the travel time to work, which might be due to the fact that they have less time constraints in compare of married couples and households with kids.

7.6 Conclusion and discussion

This chapter has aimed to enhance the limited understanding of work satisfaction concept and how this major life domain is affected by socio-demographic characteristics of individuals as well as the characteristics of one’s job. Based on the bottom up approach, path analysis method was used to capture the links between various variables and their impacts on overall work satisfaction. The findings highlight the major influence of perception regarding the social environment, work quality and privacy on feeling satisfied with work. The comparative impact of work quality versus salary should be a reminder that engaging in meaningful work dominates over salary. Moreover, the study shows that bad working space restricts individuals to show their abilities and thereby reduce the overall work satisfaction from different sides.

This study has some limitations in going into more details. First, unlike other chapters (housing, neighborhood and transport) in this chapter we could not use the whole sample due to the fact that some of the samples were unemployed, retired or student. Therefore, the data on which the analysis is based provides a small sample of people’s work satisfaction. Second, because of the psychological nature of the work satisfaction topic more psychological measures are required, which is beyond the scope of this project. And third, most of the studies on work satisfaction just explore some dimensions and usually ignore factors such as commute to work. Therefore, finding the supporting literature with comprehensive assessments was challenging.

Although the findings presented in this chapter are interesting, it is based on the bottom-up approach, while a reverse direction (top-down approach) is also conceivable, reflecting how overall happiness with work can spill over to other bottom factors. For instance, as Olsson et al. (2013) argued, the happiness of having a job after being unemployed for a long time can spill over to satisfaction with work commute. Clearly, this study and these samples were not designed to address these issues in the first place, but these insights may provide many opportunities to expand future research.

142

Quality of Urban Life (QOUL)

Quality Of Urban Life (QOUL)

8.1 Introduction

As the objective of this research project is investigating Quality Of Urban Life (QOUL) with an explicit focus on the role of the physical and built environment, in the last four chapters, four main domains (housing well-being, neighborhood well-being, transport well-being and job well-being) were examined in detail. For this chapter, the initial idea was to include all the attributes that significantly contribute to the estimations in the last four chapter, however, complexity of the model and the large number of variables (around 100 significant variables) including the small size of data made software unable to estimate the model. Muthén & Muthén (2002), mentioned that 5 to 10 cases is needed per parameter, so to estimate such a model we needed minimum 500 respondents.

143

Quality of Urban Life (QOUL)

Due to the computational issues, we decided to include just the overall satisfaction of domains to the final model that is presented in this chapter. So, following our domain-based satisfaction approach, all other main life domains viz. income well- being, health well-being, family life well-being, social well-being and leisure well-being will be introduced and included in the final assessment of well-being. A bottom-up approach, which is common in quality of urban life studies (e.g. Marans & Rodgers, 1975; Campbell et al., 1976; Sirgy & Cornwell, 2002) is used in our analysis, where well-being in different life domains is used to predict overall QOUL.

In formulating the conceptual framework of this thesis (see chapter 2), apart from the nine domains contributing to overall QOUL, socio-demographic variables may also influence it. Thus, this chapter examines overall QOUL by using the well-being of the selected life domains, and also controlling for socio-demographic variables. Explicitly, the following research questions will be explored:

IV. What are the main predictors of QOUL?

V. To what extent do these domains contribute to predicting QOUL?

VI. What are the relationships among these domains and to what extent do they affect each other?

8.2 Literature review

As discussed earlier in chapter 2, the conceptualization of this research project is motivated by the life domain approach and particularly the bottom-up spillover theory. Thus, in this chapter we focus on the QOUL studies that have applied this approach.

8.2.1 Domain-based QOL

Researchers have achieved considerable advances in methodological and theoretical approaches of QOL, and thus the field has witnessed new perspectives and paradigms. Such advances have made QOL a field in transition, which motivates researchers to go beyond the accepted methods and examine new domains (Ryan & Deci, 2001). Domain- based or life domain approach in QOL studies indicates that the relationship between a person’s subjective well-being in different areas of life contributes to the explanation of his/her overall life satisfaction (Delhey, 2014).

144

Quality of Urban Life (QOUL)

The life domain approach in QOL research is not a new approach, as Sirgy (2011) stated. Andrews and Withey (1976) and Campbell et al. (1976) are the pioneers in using life domain approach for QOL research. They examined satisfaction with various life domains such as satisfaction with family life and work to explain life satisfaction as a whole. Later, Schalock (1989) identified seven main domains that significantly influence QOL. Moreover, Cummins (1996) reviewed 1500 articles related to QOL with special focus on the life domain approach and found 351 different domain names. More recently, Van Praag et al. (2004) investigated the effects of seven most important life domains on well-being, and they listed work well-being, economic well- being, housing well-being, health well-being, leisure well-being, marital well-being and social well-being as the main life domains. However, a big portion of domain-specific studies focuses on satisfaction with particular life domain in relation to subjective well- being. For instance, job satisfaction studies or health well-being studies. Also, many studies just focus on individuals’ conditions in one domain and its relation with well- being, such as studies on unemployment and well-being or physical disabilities and well- being. Despite the general acceptance of such relationships between satisfaction in life domains and overall life satisfaction, the nature of casualty has raised questions and still is a subject of debate. Some researchers argue the existence of reversed causality, which advocates the influence of overall life satisfaction on satisfaction in other life domains (e.g. Argyle, 2013). This is called the “Top-down approach” and clearly points to the dominant influence of ones’ personality on evaluating life domains and life as a whole.

8.2.2 Factors determining QOL

Research on subjective well-being have shown that when people evaluate their life, they review their feelings regarding various life domains and based on the relative importance of those domains sum these evaluations to make a final judgment. This whole process is influenced by individuals’ socio-demographic characteristics, which reflect one’s life conditions. Hereafter, some of the most important socio-demographic variables and life domains are discussed in more detail.

8.2.2.1 Individuals’ socio-demographic characteristics

Among the socio-demographic variables, age has been always considered as one of the important characteristics of individuals, which influence well-being. As age is related to maturity that provides stability of emotions and to some extent positiveness, it is said 145

Quality of Urban Life (QOUL) that life satisfaction my increase with age. Conversely, health issues of elderly can have negative effects on their well-being. Thereby, in general age has many physiological and psychological implications for QOL, which should be considered in any QOL research (Orth et al., 2010). The evidence of the influence of gender on QOL is conflicting. Dolan et al. (2008) has argued that some authors found that females have higher levels of QOL, but the evidence at large has not indicated any gender differences in average subjective well- being. McCrea et al. (2005) also asserted that the importance of different life domains in predicting QOL does not differ among males and females.

The findings of empirical research suggest that marriage has a positive influence on feeling positive (e.g. Blanchflower & Oswald, 2004). The benefits of marriage are listed as feelings of attachment, stress buffering and social support. However, these benefits are related to happy marriage and those who are not happy with their partner reported lower level of QOL. In general, as Diener (2009) argued, the trends of marriage, divorce and their impact on QOL are consistent around the world. In terms of the influence of education on QOL, the findings are conflicting (Dolan et al., 2008). On one hand, some researchers believe that there exists a positive relationship, as education may help reaching life goals (e.g. Blanchflower & Oswald, 2004). On the other hand, higher education can rises one’s expectations, which in turn can negatively influence subjective well-being (Chevalier & Feinstein, 2006). Regarding employment status, most research findings reveal that unemployment has a negative effect on subjective well-being (e.g. Van der Meer, 2014; Clark et al., 2015). Research has also shown that loss of job is not just about loss of income, but it affects other source of well-being such as social approval (Van der Meer, 2014).

It is a common belief that money can buy happiness. However, the relationship between income and life satisfaction has been one of the most controversial topics in the literature on well-being (e.g. Ferrer-i-Carbonell, 2005; Boyce et al., 2010). Some studies show a weak correlation among income and subjective well-being (e.g. Boyce et al., 2010) while few studies provide evidence that higher-income people experience much higher average well-being. Therefore, it seems that the debate is more on the size of the effect rather than the existence of effect. Regarding the influence of ethnicity status on QOL, as expected, the findings are also conflicting. For instance, Safi (2009) revealed that being an immigrant negatively influence subjective well-being. In contrast, Bartram (2010) has shown that immigrants tend to earn more

146

Quality of Urban Life (QOUL) income, which in turn increases their life satisfaction. Although, people immigrate to enhance their QOL, however, there are many other factors that influence satisfaction feeling such as integration in society.

8.2.2.2 Satisfaction with key life domains (life domains well-being)

Income well-being is one of the three most discussed topics (along with health and happiness) in QOL research. Having adequate amount of income is necessary for well- being. So, the common belief is that, the primary condition for income well-being is whether income can fulfill the basic needs. However, how much income can satisfy people has remained a question, as income well-being is dependent on many aspects, such as personality of individuals and material aspirations, as well as the context of living and reference points for comparison (Easterlin et al., 2011). Family life well-being has proven to have a strong correlation with overall life satisfaction. Psychologists emphasize on the role of sense of belonging to family in having positive feeling about life (Dolan et al., 2008). Satisfaction with marriage and relationships within the family are important predictors of family life well-being. Evidence also suggests that family life well-being may influence feelings in other life domains (Sirgy, 2012).

Social well-being is another important life domain with strong influence on QOL (e.g. Diener & Seligman, 2002). It refers to quality and quantity of social interactions and participating in community. Moreover, the support and care ones can receive from others play a significant role in feeling satisfied with social life (Hahn et al., 2010). Satisfaction with health or health well-being is found to be an essential factor contributing to an individual’s subjective well-being, and actually some authors found it as the most important determinants of well-being (e.g. Campbell et al., 1976). It is defined as the perceived personal health and may influence all life domains substantially. Diener (2009) asserted that the impact of objective health on well-being is weak compared with subjective health as perceived health may cover more factors related to mental health.

There is a lot of evidence that suggests leisure well-being plays a significant role in life satisfaction (e.g. Chen et al., 2010). Leisure well-being refers to the extent that individuals feel alive and sets goals for their free time (Wang et al., 2011). Leisure activities have many forms, apart from the home-based and routine leisure activities, going on vacation is one of the main forms of leisure activities with strong influence on well-being. Few studies have explicitly examined the link between travel trip satisfaction

147

Quality of Urban Life (QOUL) and well-being and found a significant correlation among them (e.g. Dolnicar et al., 2012).

Job satisfaction or career well-being has also been listed as one of the main life domains, and many empirical studies have shown a strong association among work well-being and overall life well-being (e.g. Lee et al., 2015; Senasu & Singhapakdi, 2017). In most cases, job is a source of income and provides necessary facilities for living, however, at a higher level, career is related to one’s goals, desires and achievements. From this perspective, job is more than just an income source, but can shape identity and influence personal growth (Hulin, 2002). Thus, it is a rewarding activity and pleasure of doing desirable work is a significant part of life satisfaction.

Several empirical studies found a direct link between quality of life and housing well-being, which resulted in the recognition of housing well-being as a criterion for determining QOL (e.g. Fernández-Carro, et al., 2015; Fernández-Portero et al., 2017; Zebardast & Nooraie, 2018; Zhang et al., 2018). Housing is a basic need and for most people house is the most important place life experience happens. Moreover, house well-being may influence family well-being as well as social interactions and home based leisure activities. Thus, the influence of housing well-being on other life domains is also profound.

The central position of neighborhood in providing and facilitating life with different elements explains why neighborhood well-being can be a significant predictor for QOL (Austin et al., 2002). Because of this important role, many researchers have explored the associations between various neighborhood characteristics, residents’ satisfaction and well-being (e.g. Cao, 2016; Zhang & Zhang, 2017). Furthermore, neighborhood well-being may influence other life domains such as housing well-being and social well-being.

In recent years, a limited number of studies focused on the link between travel satisfaction and well-being (e.g. Sirgy et al., 2011; Ettema et al., 2011; Friman et al., 2013; Aydin et al., 2015; Gao et al. 2017). Clearly, a good transport system can influence different components of everyday living positively, whereas inefficient transport can be costly, time consuming, risky and stressful for commuters, and results in a huge loss of abilities and resources. Most of them found a significant but indirect link between travel well-being and subjective assessments of life. However, based on including the personality trait the direction of effect may be changed. Moreover,

148

Quality of Urban Life (QOUL) transport well-being is directly associates with neighborhood assessments and the link has extensively examined in many empirical studies (e.g. De Vos et al., 2016).

8.3 Analysis and results: Multiple regression analysis

In order to explore the implicit associations between well-being in various life domains and residents’ overall assessment of their well-being, a multiple regression analysis was conducted with overall QOUL as the dependent variable and 18 independent variables. From these independent variables, nine of them are socio-demographic variables and the rest are well-being ratings in various life domains. We assumed that the scales, which used to measure well-being have interval properties (same degree of difference between items) based on many studies that utilized such scales (e.g. Diaz- Serrano & Stoyanova, 2010).

Since one of our regression assumptions is that the residuals (prediction errors) are normally distributed, therefore, checking the histogram of residuals allows us to check the extent to which the residuals are normally distributed. The well-behaved residuals can lead to correct p values and thus are very important for any linear regression analysis. Our histogram (Figure 8.1) shows a fairly normal distribution. Thus, based on these results, the normality of residuals assumption is met. The other important issue to check is multicollinearity as it can reduce the reliability of the model, and also causes large standard errors and incorrect coefficient values. Choosing this option will add two new columns to the coefficient table (Tolerance and VIF). The tolerance value is between 0 and 1, and a low tolerance value for a variable means a strong relationship between this variable and the other variables. VIF is just the reciprocal of tolerance, so the smaller value of it, indicates the less risk for multicollinearity (Brace et al., 2006). A commonly used rule of thumb is that a VIF higher than 10 indicates multicollinearity (Cohen et al., 2014). Therefore, based on tolerance and VIF values in the collinearity column of coefficient Table 8.2, none of the predictor variables have strong relationship with each other.

Furthermore, the normal P-P Plot of regression residuals (Figure 8.2) can also be used to assess the assumption of normally distributed residuals. If data points (little circles) follow the diagonal line, the residuals are normally distributed. Our P-P plot shows minor deviation and thus the normality of residuals assumption is satisfied. The results about R-squared (R2) and adjusted R-squared are shown in Table 8.1. It shows that the indicators accounted for almost 61 percent of variance of QOUL.

149

Quality of Urban Life (QOUL)

Figure 8.1 The Histogram of the overall QOUL

Figure 8.2 Normal Probability Plot (P-P) of the regression standardized residual dependent variable for Overall QOUL

150

Quality of Urban Life (QOUL)

Table 8.1 Goodness-of-fit of the regression model for overall QOUL

R R2 Adjusted R2 Std. Error of the Estimate

.780 .608 .571 .776

Table 8.2 The estimated parameters of the overall QOUL using multiple regression analysis

Unstandardized Standardized Collinearity

Coefficients Coefficients Statistics

B Std. Error Beta t Sig. Tolerance VIF

(Constant) .812 .798 1.018 .310

Age -.002 .004 -.032 -.570 .569 .664 1.506

Education .102 .139 .035 .738 .461 .898 1.114

Male .020 .114 .008 .172 .863 .896 1.116

Single -.020 .152 -.008 -.131 .896 .559 1.788

Household size .022 .060 .022 .372 .710 .575 1.738

Unemployed -.234 .211 -.063 -1.109 .269 .636 1.572

Family loss -.144 .130 -.053 -1.113 .267 .918 1.089

Household income -.041 .033 -.061 -1.244 .215 .864 1.157

Dutch .403 .213 .098 1.889 .060 .767 1.303

Family life well-being .045 .035 .078 1.280 .202 .555 1.802

Transport well-being .004 .058 .004 .072 .943 .676 1.479

House well-being .129 .055 .145 2.328 .021 .533 1.875

Job well-being .032 .017 .102 1.834 .068 .664 1.507

Neighborhood well- .152 .068 .141 2.219 .028 .508 1.968 being

Social well-being .236 .054 .231 4.386 .000 .747 1.339

Health well-being .064 .042 .088 1.505 .134 .604 1.655

Income well-being .134 .036 .242 3.677 .000 .475 2.105

Leisure well-being .069 .039 .123 1.775 .078 .432 2.315

151

Quality of Urban Life (QOUL)

8.3.1 Interpretation

Regression coefficients in this analysis represent the magnitude of the influence of the explanatory variables on the residents’ overall well-being (Table 8.2). Among all of the indicators in this analysis, four variables show a significant relationship with the overall QOUL (P<.05). None of the socio-demographic variables is among the significant predictors in explaining Overall QOUL. “Neighborhood well-being”, “House well-being”, “Social well-being” and “Income well-being” significantly and positively influences “Overall QOUL”. The greatest coefficients in the regression model belong to “Social well- being” (.23). This mainly indicates that overall QOUL is highly influenced by satisfaction with social life. The second large impact belongs to “Neighborhood well-being” (.14) followed by “Income well-being” (.24) and “House well-being” (.15).

Although multiple regression analysis provides some insights into the impact of socio-demographic characteristics and well-being in all the mentioned life domains on overall QOUL, but this method is unable to capture the indirect relationships and the complex associations among these variables. Thereby, further analysis is necessary to understand the mechanism of effects among the study variables, which will be explained in the next part.

8.4 Analysis and results: Path analysis

The key objectives of this chapter were to evaluate the proposed theoretical model (Chapter 2) and to understand the interrelationships among the study variables in formulating QOUL. A path analysis is deemed to be an adequate solution for the achievement of these objectives, thus is used in this chapter. By this method, it is possible to conduct simultaneous assessments of the associations among multiple dependent and independent attributes, which can increase the statistical efficiency (Hair Jr et al., 2016).

8.4.1 Conceptualization

Building on the aforementioned logic from the domain-based studies of QOL and the relevance of these studies to investigate the link among different life domains, and socio-demographic variables in predicting overall QOUL, a conceptual model is elaborated (Figure 8.3). In essence the model consists of a series of link among satisfaction measures of important life domains, vis. Family life well-being, Health well- being, Social well-being, Income well-being, Job well-being, Leisure well-being,

152

Quality of Urban Life (QOUL)

Neighborhood well-being, Housing well-being, and Travel well-being and QOUL, which can be influenced by a range of individual and household characteristics. The directions of all main paths are from life domains to QOUL, as we use a bottom-up approach. Also, it is assumed that the level of well-being in one domain may affect well-being in other domains, especially the neighboring domains (based on horizontal spill over theory). As some major life events can have long-term impacts on well-being, a variable called “Family loss” added to the model indicting whether a respondent has recently lost someone close in family.

8.4.2 Path analysis model for QOUL

Figure 8.3 The conceptual model

153

Quality of Urban Life (QOUL) and “Dutch” were insignificant, and thereby, excluded from the model. However, all the variables regarding life domains well-being are remaining in the model.

The results of the final path model (Table 8.3) demonstrate that the model excellently fit to the data. Chi Square divided by the model degrees of freedom (/Df or CMIN/DF) is an essential measure, and is equal to 1.32 for our model. In addition to that, the Root Mean Square Error of Approximation (RMSEA) should be less than .08 and ideally less than .05 to represent a great fit (Golob, 2003; Washington et al., 2010), and in our model RMSEA value is .039. Other indicators of a good fit as mentioned by many researchers (e.g. Moss, 2009) are indices that compare target model with null model, such as Comparative fit index (CFI), Incremental fit index (IFI), Tucker Lewis index (TLI), Normed fit index (NFI) and (GFI) Goodness of Fit Index. If these indexes exceed the amount of .95, then the model tends to generate a great fit. Checking all of these indicators shows that our model perfectly fits. Moreover, R squared (R2) of the ultimate endogenous variable (called squared multiple correlation value in AMOS) equals to .58, which means our proposed theoretical model satisfactorily accounts for 58 percent of the variance in residents’ overall QOUL. This moderate level of explained variance in QOUL indicates that other attributes not included in the model (such as personality) also contribute to overall QOUL.

Table 8.4 presents all the direct, indirect and total effects of the paths in the model including P and T value for each path. Our results show that statistically significant relationships were found among the main model variables as below:  All the variables in the model directly and indirectly influence QOUL.  “Income well-being” has the largest total effect on QOUL (.55), and half of its impact is indirect via its influence on other life domains.  The second largest positive effect on QOUL is the “Health well-being” effect (.27), following by “Social well-being” effect (.25).  From the socio-demographic variables “Unemployed” is the only variable, which directly correlates with QOUL.  Three variables influence overall QOUL negatively, viz. “Unemployed” (- .31), “Family loss” (-.04), and “Single” (-.02).  From the domain well-being variables, only “Transport well-being“ and “Leisure well-being” do not show a direct influence on overall QOUL.

154

Quality of Urban Life (QOUL)

Table 8.3 Goodness of the fit of the model

Fit index Estimate

Chi-square (CMIN) 86

Degrees of freedom (DF) 65

Chi-square / degrees of freedom 1.32

Root mean square error of approximation (RMSEA) .039

CFI .97

IFI .98

TLI .96

GFI .95

NFI .91

 From the socio-demographic variables, “Age” has the largest influence (.11). Its effect is slightly more than “Household income” effect (.10). The other important results that can be extracted from the model can be summarized as below:  Life domains’ well-being variables are not only influenced by socio- demographic variables, but they also have correlation with each other.  “Unemployed” has the largest influence on “Income well-being” (-.35), and also has negative influence on well-being in other life domains.  “Leisure well-being” and “Transport well-being” both are highly influenced by “Income well-being” and “Health well-being”.  “Neighborhood well-being” is highly influenced by “Health well-being” and “Transport well-being”.  “Single” only correlates with “Family life well-being”, but with negative coefficient.  “Household income”, “Age” and “Unemployed” have significant effect on “Income well-being”.  “Health well-being” is negatively influenced by “Unemployed”, “Family loss” and “Age”.  “Family loss” negatively impacts well-being of almost all life domains (except “Job well-being” and “Income well-being”).  “Age” and “Job well-being” have significant relationship with a negative coefficient.

155

Quality of Urban Life (QOUL)

Table 8.4 Standardized path estimates of direct, indirect and total effects

Dependent Path Direct Indirect Total T P

variable effect effect effect

Household income  QOUL .00 .10 .10 -- --

Unemployed QOUL -.09 -.22 -.31 -1.84 *

Age  QOUL .00 .11 .11 -- --

IW  QOUL .27 .27 .54 4.62 ***

Family loss  QOUL .00 -.04 -.04 -- --

Health Well-being  QOUL .12 .15 .27 2.15 ** Quality of TW  QOUL .00 .11 .11 -- -- Urban life NW  QOUL .13 .10 .23 2.41 **

(QOUL) LW  QOUL .00 .14 .14 -- --

HW  QOUL .14 .08 .22 2.48 ***

Single  QOUL .00 -.02 -.02 -- --

JW  QOUL .10 .00 .10 2.21 **

FLW  QOUL .14 .00 .14 2.87 ***

SW  QOUL .25 .00 .25 4.99 ***

Household income  SW .00 .05 .05 -- --

Unemployed  SW .00 -.13 -.13 -- --

Social Well- Age  SW .00 .13 .13 -- --

being IW  SW .00 .26 .26 -- --

(SW) Family loss  SW .00 -.03 -.03 -- --

Health Well-being  SW .09 .11 .21 1.32 *

TW  SW .00 .08 .08 -- --

NW  SW .00 .14 .14 -- --

LW  SW .24 .00 .24 3.38 ***

HW  SW .30 .00 .30 4.78 ***

Household income  FLW .00 .06 .06 -- --

Unemployed  FLW .00 -.12 -.12 -- --

Age  FLW .00 .09 .09 -- --

IW  FLW .00 .35 .35 -- --

Family Life Family loss  FLW .00 -.04 -.04 -- --

Well-being Health Well-being  FLW .12 .16 .28 2.00 ** (FLW) TW  FLW .00 .05 .05 -- --

156

Quality of Urban Life (QOUL)

LW  FLW .53 .00 .53 8.52 ***

Single  FLW -.14 .00 -.14 -2.62 ***

Household income  JW .00 .05 .05 -- --

Job Well- Unemployed  JW .00 -.09 -.09 -- --

being (JW) Age  JW -.38 .04 -.34 -6.03 ***

IW  JW .25 .00 .25 3.95 ***

Household income  HW .00 .06 .06 -- --

Unemployed  HW -.11 -.10 -.21 -1.88 **

House Well- Age  HW .21 .09 .30 3.78 ***

being (HW) IW  HW .18 .12 .30 2.97 ***

Family loss  HW .00 -.02 -.02 -- --

Health Well-being  HW .00 .13 .13 -- --

TW  HW .00 .19 .19 -- --

NW  HW .45 .00 .45 8.01 ***

Household income  NW .00 .05 .05 -- --

Neighborhood Unemployed  NW .00 -.09 -.09 -- --

Well-being Age  NW .14 .00 .14 2.44 **

(NW) IW  NW .00 .26 .26 -- --

Family loss  NW .00 -.041 -.04 -- --

Health Well-being  NW .23 .07 .30 3.71 ***

TW  NW .42 .00 .42 6.90 ***

Household income  TW .00 .07 .07 -- --

Transport Unemployed  TW .00 -.12 -.12 -- --

Well-being Age  TW .00 .03 .03 -- -- (TW) IW  TW .27 .08 .35 3.66 ***

Family loss  TW .00 -.02 -.02 -- --

Health Well-being  TW .16 .00 .16 2.22 **

Household income  .00 .09 .09 -2.13 **

Health Well-being

Health Unemployed  Health Well- .00 -.18 -.18 -- -- being HW Well-being Age  Health Well-being -.13 .08 -.05 -- --

IW  Health Well-being .50 .00 .50 8.33 ***

Family loss  Health Well- -.14 .00 -.14 -2.31 ** being

Income Household income  IW .19 .00 .19 3.02 ***

157

Quality of Urban Life (QOUL)

Well-being Unemployed  IW -.35 .00 -.35 -5.65 ***

(IW) Age  IW .15 .00 .15 2.47 ***

Household income  LW .00 .10 .10 -- --

Unemployed  LW .00 -.19 -.19 -- --

Leisure Age  LW .14 .04 .18 2.44 ***

Well-being IW  LW .35 .18 .53 5.46 *** (LW) Family loss  LW .00 -.04 -.04 -- --

Health Well-being  LW .29 .02 .31 4.63 ***

TW  LW .10 .00 .10 1.68 *

Note: ***,**,*  significance at 1%, 5%,10% level.

 “Neighborhood well-being” shows a significant effect on “House well- being”.  “Health well-being” influences all life domains except “Income well- being”.  “Transport well-being” positively impacts “Neighborhood well-being” and “House well-being”.  “Leisure well-being” shows a significant effect on “Family life well- being” and “Social well-being”.  “Family life well-being”, “Social well-being” and “Job well-being” are the only domains with just direct influence on overall QOUL.

8.5 Interpretation

Our results show that the level of well-being in various life domains significantly contributes to the overall QOUL (Figure 8.4). Moreover, most of the socio-demographic attributes indirectly affect the QOUL assessments. To have a better overview regarding the contribution of all the variables in our estimated model, below they are discussed in more details.

8.5.1 The role of socio-demographic attributes on QOUL

As mentioned, most of socio-demographic variables indirectly affect QOUL. Unemployment, however, is an exception and has a substantial negative and direct influence on QOUL. This direct effect implies the importance of having a job on one’s life.

158

Quality of Urban Life (QOUL)

The other interesting result is related to the impact of unemployment on income well- being and house well-being, which reflects the financial side of unemployment rather than the other possible negative effects (such as social effects). Therefore, based on our findings the main negative influence of unemployment is the financial burdens that it imposes to individuals. Age is another significant variable with positive influence on QOUL. Regarding other associations of age, as expected, older people have lower satisfaction with their health situation, while income well-being, house well-being and neighborhood well-being increase with aging (e.g. Parkes et al., 2002; Chapman & Lombard, 2006). However, the negative correlation between age and job well-being contradicts some previous findings of studies including our job well-being section (see chapter 8), which reveals positive influence of age on job well-being.

Our findings also indicate that being single is significantly associated with reduced family life well-being, which in turn influence QOUL negatively. McCrea et al. (2005) also found that single people have a significantly lower overall life satisfaction than couples and families with children. As expected, family loss is negatively correlates with health well-being, and via that, it negatively influence well-being in almost all the life domains (except job well-being and income well-being). This indicates that the grief of losing someone in family can affect all aspects of life including health (mind and body), relationships and in general the whole feeling towards well-being. Regarding household income, the link between household income and QOUL is indirect and through income well-being. Similarly, well-being in all other life domains is influenced by household income indirectly. However, many authors (e.g. Kahneman & Deaton, 2010) have argued that low income reduces life satisfaction, but as long as basic needs have been met. This implies that further rises in income are not associated with further increases in well-being.

8.5.2 The influence of life domains on QOUL

The findings of this study assert that the most important predictor of QOUL is income well-being followed by health well-being and social well-being, respectively. Satisfaction ratings regarding physical domains such as neighborhood well-being and house well- being have less impact compared with income well-being, health well-being and social well-being domains. Furthermore, job well-being and transport well-being have the least total effect on QOUL. The dominant role of income well-being on QOUL clearly reflects

159

Quality of Urban Life (QOUL)

sis for sis QOUL

Results of path analy of path Results

8 Figure

160

Quality of Urban Life (QOUL) the importance of the financial aspects in maintaining a good life. Positive assessment of income is associated with all life domains directly and indirectly, and thereby has a huge impact on QOUL.

It is good to note that most researchers believe that expectations and income aspirations determine income well-being rather than the absolute income, while the basic needs fulfilled (Frey & Stutzer, 2005). Furthermore, the influence of income well- being on other life domains is remarkable, particularly on non-physical domains such as health well-being. Health well-being influences almost all the domains directly and indirectly. Different research reveals that there is a robust link between physical health, mobility and activity satisfaction, therefore, quality and quantity of commuting, leisure activities, and social activities are influenced by health well-being (Chang et al., 2016).

However, perceived health deteriorates with age and family loss. Our findings also uncovered an indirect and negative correlation between unemployment and health well- being, which has also been reported in other studies. As the financial burden imposes by unemployment to one’s life can significantly reduce health well-being.

Individuals’ sense of social well-being plays a significant role in QOUL, and our findings also show a robust link among them. We can also observe that better health can promote social well-being. Moreover, the correlation that we obtained among leisure well-being and social well-being has been also admitted by other research findings. For instance, Chen et al., (2010) advocated that many forms of leisure activities are linked with social interactions, which can fulfill variety of social needs. Other research findings (e.g. Cornwell, 2002; Bjornskov, 2003; Dolan et al., 2008; Sirgy et al., 2010) also corroborate the relationship between house well-being and social well- being, as certain housing amenities associates with the ability of hosting, entertaining and socializing with friends and relatives. We can see from the path coefficients that neighborhood well-being and house well-being significantly contributes to QOUL, which is also in line with the findings of other empirical research (e.g. Bjornskov, 2003; Dolan et al., 2008; Marans, 2012). Moreover, regarding the link among neighborhood well- being and house well-being, it has been shown that positive assessment of neighborhood directly influence housing well-being, as location of housing is attached to the overall feeling regarding house (e.g. Teck-Hong, 2012). Based on our results, age and income well-being positively contributes to housing well-being, while unemployment affects it negatively. The fact that no direct relationship is observed between transport well-being and QOUL is also in line with some previous research

161

Quality of Urban Life (QOUL) findings (e.g. Currie et al., 2010; Stanley et al., 2011). Travel well-being, however, influences neighborhood well-being and leisure well-being and via them indirectly impacts QOUL. But as expected, its contribution to QOUL is less than house and neighborhood well-being.

Regarding the effect of family life well-being on QOUL, our results show a direct link among them. Generally, family life is a strong predictor of well-being and negatively influenced by family loss and unemployment, which is also reflected in our findings. Also, as expected, married people are more satisfied with their family life than single people. Finally, our results also indicate that higher levels of job well-being promote QOUL. Based on our findings in job well-being chapter (see Chapter 7) satisfaction with quality of work has a major effect on job well-being. This reflects that the need for doing a meaningful activity plays a key role in personal growth and well- being, and job well-being can fulfill higher levels of needs.

8.6 Conclusions and discussion

This chapter presents a methodological and empirical contribution to the literature on the residents’ QOUL using path analysis. Path analysis allows analyzing multiple domains of QOUL and the links among them simultaneously. It also enables us to have a better understanding of the complexity of QOUL concept. We used the most important domains including the socio-demographic characteristics of individuals to conceptualize QOUL. Our findings highlight the major and distinct influence of the income well-being, health well-being and social life well-being on overall QOUL. Moreover, from the socio- demographic attributes, age, income, marital status and employment status significantly contribute to QOUL. However, no significant difference is observed in the QOUL of males and females, highly educated and low educated as well as Dutch and minorities. The other noteworthy finding of this research is the associations among various life domains, which is called “Horizontal spillover” (Sirgy, 2012), and explains why extreme dissatisfaction or lack of satisfaction in one life domain may influence (spill over in) other life domains.

In general, quality of life and well-being are complex notions. On one hand, life domains evaluations are grounded on objective realities, and on the other hand, these judgments are influenced by personality and psychological moods of individuals. In this regard, optimism and positiveness can play a key role in subjective well-being of individuals. Also, there is much overlap among the various domains of well-being and in

162

Quality of Urban Life (QOUL) some cases the relationships are two-fold, particularly between the neighboring domains, such as house well-being and neighborhood well-being or neighborhood well- being and transport well-being. These inter domains interactions cause the interpretation more challenging. Moreover, QOUL studies can address microlevel or macrolevel well-being. While, this study has focused on microlevel subjective well- being, macrolevel studies should focus at country-level political and economic circumstances, as well as community’s stability, social cohesion and culture.

Although this chapter has examined all the key domains affecting microlevel QOUL, but considering the numerous factors and broadness of the concept, there are many opportunities to expand research in this area. The applications of such studies are also extensive and can potentially address policy makers and urban planners as well the residents who are curious about themselves and their living environment.

163

Conclusions and future work

164

Conclusions and future work

Conclusions and Future work

9.1 Summary

Despite extensive research on quality of life over the past half century and the long history of urbanization in human civilization, the study of quality of life in urban places has only recently received attention. This recent attention is a response to concerns about various issues in cities, and the increasing threats over human life. Moreover, it reflects a shift in urban planning away from basic needs of citizens to a larger spectrum of policy goals. In this regard, quality of urban life (QOUL) studies can provide practical solutions to some of these challenges and provide guidelines and strategies for policy makers and urban planners to enhance subjective well-being of residents.

165

Conclusions and future work

Previous studies mostly focused on either specific issues in a specific urban life domain such as housing, or just depicted a general picture of correlates of QOUL. However, these studies did not consider the corresponding effects among various life domains, while these effects may be significant and cannot be ignored. Also, the strength of the link between various attributes of life domains cannot be fully captured just by considering the objective characteristics of the environment, but the subjective assessments that stem from individuals conditions and constraints should be addressed as well. This PhD research tries to address these issues and fill these gaps through an integrative approach considering the corresponding effects among various life domains and urban life characteristics. Here, the assumption is that physical characteristic of the environment can potentially affect the subjective assessment of residents in various concrete areas of life, which can be classified into few main domains. In formulating the conceptual model of this study, overall QOUL was assumed to be a function of the person-place-activity relationship, and thereby, the main urban life domains emerged from this concept. Regarding the place concept, housing well-being and neighborhood well-being were included in the model, as both house and neighborhood are key living places. In terms of main urban activities, work well-being and transport well-being domains were added to the model, as both may significantly influence QOUL. So the main four domains of housing well-being, neighborhood well-being, transport well- being, and job well-being were selected as main urban life domains and investigated in more details. The neighborhood domain includes the opportunity the environment offer to conduct daily activities. In the following, the main results of each chapter will be presented.

Chapter 4 explored the effects of both objective and subjective attributes of housing on well-being, using respectively a regression model and a path model. The empirical results of the path analysis provided evidence that subjective attributes such as satisfaction with house ownership, house type, house design, house exterior space and safety have significant effect on housing well-being. Among them, house type satisfaction shows the highest influence on house satisfaction followed by house design satisfaction. Among the objective variables, housing cost, ownership and safety facility are the dominant attributes in influencing housing well-being, as it is observed that low- cost house, rental house and lack of security facilities indirectly and negatively influence housing well-being. It is also good to note that socio-demographic variables, household income and length of residency show indirect and positive effect on housing well-being.

166

Conclusions and future work

Chapter 5 explicitly focused on neighborhood satisfaction as the neighborhood is the second immediate space after housing that urban residents encounter during their life course, playing a key role in fulfillments of residents' daily needs. To understand the complex relationship among the various attribute of neighborhood well-being domain, the conceptual model of this chapter categorized the main attributes affecting neighborhood well-being into three main categories, namely neighborhood characteristics (physical, environmental and social condition), neighborhood physical activity (biking and walking), and neighborhood services (e.g. grocery shopping, sport center, etc.). To measure these constructs (latent variables) a structural equation model was developed and applied. The results highlight the major and distinct influences of neighborhood characteristics followed by neighborhood services and neighborhood physical activity. Moreover, satisfaction with social conditions has the most prominent effect on neighborhood well-being followed by satisfaction with environmental condition and physical sub-domains. In the neighborhood services category, only grocery shopping was shown to be influential compared with the other six major services in neighborhood (e.g. restaurant, school, etc.). In the neighborhood physical activity category, both walking and biking conditions were found to be significant predictors, however, walking condition shows a stronger correlation. Among the socio-demographic factors, only marital status has a significant effect on neighborhood well-being, as being single negatively correlates with neighborhood well-being.

Chapter 6 investigated transport well-being (travel satisfaction), which is a multidimensional and dynamic concept. To identify the main drivers of commuters’ travel satisfaction, travel behavior of residents should take into account. Thereby, information regarding respondents’ trips during peak and off-peak hours, and their travel mode choice were obtained. Moreover, based on the specific mode use, satisfaction with various trip attributes were rated to help identifying the main drivers of satisfaction, and the extent to which these attributes influence travel satisfaction. Using private transport mode during peak hours was explained by six attributes such as satisfaction with parking availability, safety of roads, etc., and public transport mode use was explained by seven variables such as travel time, safety, etc. The empirical results of structural equation analysis provided evidence that only travel during peak hours influences travel well-being, and consequently avoiding rush hours plays a positive and significant effect on travel satisfaction. Also, all chosen trip attributes could significantly explain the travel satisfaction assigned to that particular mode. In addition to all the travel characteristics,

167

Conclusions and future work our model also illustrated a significant effect of built environment characteristics and satisfaction with accessibility and neighborhood as influential factors. In this regard, feeling satisfied with travel time to major places (e.g. family/friends, recreational centers) plays an in important role in travel satisfaction. Accessibility clearly reflects the relationship that exists among one’s desires and obligations to do various activities, and the opportunities of the environment that can fulfill these needs. Moreover, results show that females' single people experience higher levels of travel satisfaction.

Chapter 7 investigated the role of socio-demographic variables, employment status (part-time vs full-time), and various work-related attributes in a multi-attribute structure, where satisfaction in three domains of work conditions (salary, workload, etc.), work location (travel time, safety of area, etc.), and workspace (privacy, size, etc.) were used to explain job satisfaction. The path analysis results indicated a strong influence of work conditions satisfaction, followed by workspace and work location, respectively. In the work conditions category, satisfaction with quality of work has a major effect on work conditions satisfaction. Here, the comparative impact of work quality satisfaction versus salary satisfaction can be an indication that performing meaningful work dominates over salary. In the work location category, satisfaction with travel time to work is the most important attribute, while in the workspace category, satisfaction with privacy plays a key role.

There are also some intra domains relationships that can be extracted from our model. For instance, the major effects of the attributes in the workspace category on satisfaction with quality of work may indicate that bad working space restricts individuals to show their abilities, and thereby, reduce overall work satisfaction. The findings also highlight the major and direct influence of perception of the social environment and privacy on feeling satisfied with work.

Finally, Chapter 8 investigated the role of all urban life domains discussed in previous chapters including the other major life domains (which were not the focus of this research study) on overall QOUL based on domain-specific approach. The proposed theoretical model, thereby, consisted of series of links among nine major life domains, socio-demographic variables and overall QOUL. The directions of all main paths were from life domains to QOUL, as we used a bottom-up approach. Also, it is assumed that the level of well-being in one domain may affect well-being in other domains, especially the neighboring domains (based on horizontal spill over theory). Therefore, the findings of this chapter were two-fold, showing both the influence of life domains on QOUL and

168

Conclusions and future work the inter-relationship among various life domains. In order to investigate these links, a path analysis method was developed and used. Our results indicate that the proportions of explained variance (predictive power) in overall QOUL are different for the various domains. The effects of these domains can be sorted from high to low effect as: income satisfaction, health well-being, social well-being, neighborhood well-being, house well- being, transport well-being and job well-being, respectively. Although job well-being has shown to be a less influential factor compared to the other main domains, unemployment substantially decreases the degree of well-being. Our model results also indicate that all urban life domains, except transport well-being, have a direct influence on QOUL. Although it was not assumed in our initial hypothetical model, but this result is in line with the findings of other research (e.g. Gao et al. 2017 & 2018). This indirect effect runs through the neighborhood well-being domain, which emphasizes the role of neighborhood accessibility on QOUL.

Moreover, five personal characteristics (age, marital status, household income, employment status and family loss) were found to significantly affect overall QOUL. Results indicate that younger people, single and unemployed individuals including lower income households experience lower level of satisfaction regarding their life. However, unemployment is the only variable with a direct effect on QOUL. In addition, as hypothesized, family loss has shown to have a negative influence on health well-being and in turn QOUL.

The findings on the interrelationships among the main domains are also interesting. First, the major influence of income well-being on well-being in other domains is surprising, but logically explainable, as income well-being determines access to sources and facilitates life in many ways. Moreover, it can also be seen that many neighboring domains influence each other, such as the impact of transport well-being on neighborhood well-being, or the effect of neighborhood well-being on house well- being. Clearly, positive assessment of neighborhood influence housing well-being, as location of housing is attached to the overall feeling regarding house. In addition to that, housing well-being influence social well-being as certain housing amenities associates with the ability of hosting, entertaining and socializing with friends and relatives. All in all, as the four major urban domains that we investigated have significant influence on QOUL, we can conclude that our results confirm the hypothetical model of this thesis. Also, all the major attributes that we found in Chapter 4 to 8 indirectly influence QOUL.

169

Conclusions and future work

9.2 Contribution and Managerial implications

The findings of QOL studies in general and QOUL in particular have diverse and profound implications for policy, planning and design, and can serve as a basis for optimizing the well-being of urban residents. Indeed, if the actions of decision makers are founded on the results of academic research, they are more likely to be responsive to people’s needs and preferences. Similarly, many non-governmental organizations who deal with consumer behaviour can apply the indicators and methods of QOUL studies. As the interest on QOL research within urban planning grows, and the efforts to assess and measure well-being increase, the topic attracts more attention from policy makers. Thereby, the opportunities for collaboration between scholars and the public and private sector might be increased in the future. However, irrespective of the context of QOUL research, replication of studies over time can provide many useful results in terms of changes in perception of residents regarding various dimensions of the environment. In fact, monitoring QOL in longitudinal studies is the best way to broaden our perspective regarding both subjective and objective QOL.

This study provides insightful understanding in terms of various life domains, the interconnection among these domains and also their associations with QOL. Living, working and commuting in urban environment have been studied and the results discussed respectively. Regarding housing well-being, ownership found to have a significant role. In this regard, governments tendencies and efforts can assist and enable low-income families to become homeowners. In a broader sense, the global financial crises have always had a tremendous impact on the housing market. Also, housing issues has been part of political debates and always bounded to financial and also ideological matters. However, the lack of quality of life in such debates is evident. Moreover, the other interesting result in the housing chapter is related to type of house satisfaction. The positive experience of residents in StrijpS lofts reveals that combining living and working in a studio is a popular trend for young generations who cannot afford renting separate places for living and working. This basically stems from the flexible design of these studios that provide opportunities for doing various activities. Increasing the number of these type of flats, particularly in expensive cities should be considered.

In the neighbourhood well-being chapter, our results emphasize the role of the social domain and therefore fostering residents from different background can improve collective living in an area. If municipalities and related sectors develop community-

170

Conclusions and future work based organizations and encourage them to organize activities in neighbourhoods frequently, neighbourhood activities could act as a platform to engage citizens for social work and can benefit both individuals and communities.

In the transportation well-being chapter, the dominant result was the higher satisfaction of those who avoid commuting in peak hours. Consequently, encouraging to work from home, postponing trips from peak to off-peak hours, and using active modes of transport, can be some of the solutions to promote travel well-being. Additionally, peak avoidance policies for some companies such as delivering companies can reduce traffic congestion and promote the drivers well-being.

In the job-well-being chapter, one result was related to characteristics of work space and highlighted the importance of space privacy feeling in satisfaction with work place, and consequently overall job satisfaction. As open-office layout concept grows rapidly, it is important to consider the employees’ need for concentration, and it would be good that office manager could provide more private rooms. Also, using furniture to separate spaces among employees could be another option to decrease distractions and improve privacy feeling.

In the quality of urban life chapter, the results clearly show the dominant role of income well-being and the negative influence of unemployment on overall well-being of individuals. Reducing unemployment and optimizing labour market have always been challenging tasks for governments. However, it is possible to reduce some negative effects of unemployment by employing relevant strategies. For instance, reducing social isolation of unemployed citizens by organizing free events, engaging them in volunteer jobs and also assisting them in finding jobs. There are many more results that can be driven from this study regarding the social and family well-being, however it is beyond the scope of this study to recommend particular solutions for this broad range of issues.

Although these results have been extracted from a small sample, in many ways they are likely representative of the preferences of a larger population of residents. The most important function of this research and similar research for public policy could be a shift on the focus by addressing different questions. For instance, instead of asking “How work place impact productivity?” we should ask “How work place impact work well-being?” or instead of asking “How unemployment impact economic growth?” asking “How unemployment impact QOL?”. In other words, shifting from concepts such as job performance to job well-being can provide policy makers with different guidelines for their future performances.

171

Conclusions and future work

9.3 Discussion and direction of future research

The aforementioned conclusions are based on the hypothetical framework that we designed in the beginning of this study. This framework was estimated based on the data collected from four different neighborhoods in Eindhoven. The subjective measures described in this research, and in general in most of QOUL studies are based on several assumptions that can generate bias. The main assumption is that individuals can derive their overall QOUL or satisfaction in one life domain from their daily experiences. Also, retrospective evaluations of life can be biased as they mostly reflect major and recent experiences of respondents. To reduce this bias, we asked respondents to mention if they have recently lost someone in their family. Moreover, situational factors (such as environmental conditions) and moods of respondents can bias the responses toward higher or lower ratings of subjective well-being. In order to reduce environmental condition biases, our data were collected in two seasons. There is also the limitation regarding measurement methods, and the choice of attributes and domains. Our study focuses on four urban life domains, and also includes overall satisfaction of five other main domains (financial, personal, health, social and leisure). More life domains such as religious life, political environment, etc. may be relevant to include in future research and/or in other areas. Moreover, this study is only concerned with four areas in a single city in the Netherlands, and therefore, these data only reflect particular differences in the built environment. To better understand the role of urban and built environment characteristics, further investigation of QOUL in different cities and countries is recommended.

In brief, any future research needs to address a series of challenges, in order to achieve a comprehensive understanding of the role of urban living experience on QOUL. These challenges cover collection of data, modelling and measuring urban life well- being. The first challenge is collecting larger scale data, without compromising the quality and depth of the presented method. For instance, in our study the number of working people were (130) and among them (53) were part-time employee. Therefore, in order to fully understand the impact of part-time working on overall work satisfaction a larger sample would be desirable. This also applies to the transportation domain as travel well-being is highly related to mode use, and dividing the sample based on mode use needs larger samples to presents all the modes.

The second challenge concerns the nature of the relationship between QOUL and satisfaction in the different domains of life. The issue stems from the general

172

Conclusions and future work assumption that an additive relationship exists between QOUL and satisfaction in domains of life. It means that dissatisfaction in one aspect of a domain can be compensated by other aspects. For instance, by improving interior design of a house, dissatisfaction regarding house size can be decreased. However, it is shown that the same relationships are not applicable to QOUL, and increase in satisfaction with one domain may not compensate dissatisfaction or decline in another domain. For instance, an increase in job satisfaction may not be enough to compensate decline in family life satisfaction. Thus, the additive specification (used in this research) is good in predicting QOUL, but to better understand the complex mechanism of QOUL, non-linear and non- compensatory model specifications should be tested.

The third challenge concerns the direction of causality in the QOUL model. The argument regarding bottom-up vs top-down approach should also be considered to understand the impact of overall QOUL on different domains satisfaction. This may show that QOUL impact may not be uniform across domains. Moreover, our approach in this study is a cognitive representation of QOUL concept, and examined the relative importance of urban life attributes. However, future studies can also consider affective approaches.

173

Conclusions and future work

174

References

Abdallah, S., Michaelson, J., Shah, S., Stoll, L., & Marks, N. (2012). The happy planet index: 2012 report. A global Index of Sustainable Well-being, 26. Adams, A., & Bond, S. (2000). Hospital nurses’ job satisfaction, individual and organizational characteristics. Journal of Advanced Nursing, 32(3), 536-543. https://doi.org/10.1046/j.1365-2648.2000.01513.x. Addo, I. A. (2016). Assessing residential satisfaction among low income households in multi-habited dwellings in selected low income communities in Accra. Urban Studies, 53(4), 631–650. http://doi.org/10.1177/0042098015571055. Adriaanse, C. C. M. (2007). Measuring residential satisfaction: a residential environmental satisfaction scale (RESS). Journal of Housing and the Built Environment 22(3), 287–304. AhnAllen, J. M., Suyemoto, K. L., & Carter, A. S. (2006). Relationship between physical appearance, sense of belonging and exclusion, and racial/ethnic self- identification among multiracial Japanese European Americans. Cultural Diversity and Ethnic Minority Psychology, 12(4), 673-686. http://dx.doi.org/10.1037/1099-9809.12.4.673. Aiello, A., Ardone, R. G., & Scopelliti, M. (2010). Neighbourhood planning improvement: Physical attributes, cognitive and affective evaluation and activities in two neighbourhoods in Rome. Evaluation and Program Planning, 33(3), 264-275. Albert, D. P., & Gesler, W. M. (2003). How spatial analysis can be used in medical geography. In Spatial Analysis, GIS and Remote Sensing, (pp. 19-46). Boca Raton: CRC Press. Alfonzo, M. A. (2005). To walk or not to walk? The hierarchy of walking needs. Environment and Behavior, 37(6), 808-836. Allen, T. D., Golden, T. D., & Shockley, K. M. (2015). How effective is telecommuting? Assessing the status of our scientific findings. Psychological Science in the Public Interest, 16(2), 40-68. Alshmemri, M., Shahwan-Akl, L., & Maude, P. (2017). Herzberg’s two-factor theory. Life Science Journal, 14(5), 12-16. Amérigo, M., & Aragones, J. I. (1997). A theoretical and methodological approach to the study of residential satisfaction. Journal of Environmental Psychology, 17(1), 47-57. Amole, D. (2009). Residential satisfaction in students’ housing. Journal of Environmental Psychology, 29(1), 76–85. http://doi.org/10.1016/j.jenvp.2008.05.006. Amponsah-Tawiah, K. (2013). Occupational health and safety and sustainable development in Ghana. International Journal of Business Administration, 4(2), 74-78. Andrews, F. M., Withey, S. B. (1976). Social indicators of well-being: Americans Perceptions of Life Quality. New York: Plenum press.

175

References

Aragonés, J. I., Amérigo, M., & Pérez-López, R. (2017). Residential satisfaction and quality of life. In Handbook of Environmental Psychology and Quality of Life Research. Cham: Springer. Argyle, M. (1996). The Social Psychology of Leisure. London: Penguin. Argyle, M. (2013). The Psychology of Happiness. London: Routledge. Aries, M. B., Veitch, J. A., & Newsham, G. R. (2010). Windows, view, and office characteristics predict physical and psychological discomfort. Journal of Environmental Psychology, 30(4), 533-541. Aulia, D. N., & Ismail, A. M. (2013). Residential satisfaction of middle income population: Medan city. Procedia-Social and Behavioral Sciences, 105, 674-683. Austin, D. M., Furr, L. A., & Spine, M. (2002). The effects of neighborhood conditions on perceptions of safety. Journal of Criminal Justice, 30(5), 417-427. Avila-Palencia, I., Panis, L. I., Dons, E., Gaupp-Berghausen, M., Raser, E., Götschi, T., ... & Anaya-Boig, E. (2018). The effects of transport mode use on self- perceived health, mental health, and social contact measures: a cross-sectional and longitudinal study. Environment International, 120, 199-206. Aydin, N., Celik, E., & Gumus, A. T. (2015). A hierarchical customer satisfaction framework for evaluating rail transit systems of Istanbul. Transportation Research Part A: Policy and Practice, 77, 61-81. Baum, S., Arthurson, K., & Rickson, K. (2010). Happy people in mixed-up places: the association between the degree and type of local socioeconomic mix and expressions of neighbourhood satisfaction. Urban Studies, 47(3), 467- 485. Baumeister, R. F., & Forgas, J. P. (2018). The Social Psychology of Living Well: Historical, Social and Cultural Perspectives. In The Social Psychology of Living Well (pp. 17-34). New York: Routledge. Bernardo, F., & Palma-Oliveira, J. M. (2016). Urban neighborhoods and intergroup relations: The importance of place identity. Journal of Environmental Psychology, 45, 239-251. Birnholtz, J. P., Gutwin, C., & Hawkey, K. (2007). Privacy in the open: how attention mediates awareness and privacy in open-plan offices. Proceedings of the 2007 International ACM Conference on Supporting Group Work, 51-60. Bissell, D. (2010). Passenger mobilities: affective atmospheres and the sociality of public transport. Environment and planning D, Society and Space, 28(2), 270-289. Biswas-Diener, R., & Diener, E. D. (2006). The subjective well-being of the homeless, and lessons for happiness. Social Indicators Research, 76(2), 185-205. Bjornskov, C. (2003). The happy few: Cross-country evidence on social capital and life satisfaction. Kyklos, 56, 3–16. Blanchflower, D. G. & Oswald, A. J. (2004). Well‐being over time in Britain and the USA. Journal of Public Economics, 88, 1359–86. Bloch-Jorgensen, Z. T., Cilione, P. J., Yeung, W. W., & Gatt, J. M. (2018). Centeredness Theory: Understanding and measuring well-being across core life domains. Frontiers in Psychology, 9.

176

References

Bocarejo, S. J. P., & Oviedo, H. D. R. (2012). Transport accessibility and social inequities: a tool for identification of mobility needs and evaluation of transport investments. Journal of Transport Geography, 24, 142-154. Booth, A. L. and van Ours, J. C. (2008). Job satisfaction and family happiness: the part‐time work puzzle. The Economic Journal, 118 (526), F77–F99. Bowling, N. A., Eschleman, K. J., & Wang, Q. (2010). A meta‐analytic examination of the relationship between job satisfaction and subjective well‐being. Journal of Occupational and Organizational Psychology, 83(4), 915-934. Boyce, C. J., Brown, G. D. A., & Moore, S. C. (2010). Money and happiness: Rank of income, not income, affects life satisfaction. Psychological Science, 21(4), 471- 475. doi: 10.1177/0956797610362671. Brace, N., Kemp, R., Snelgar, R., & Lee, J. (2006). SPSS for psychologist: A guide to data analysis using SPSS for windows. Hampshire: Macmillan publishers limited. Brannon, D., Barry, T., Kempster, P., Schreiner, A. & Vasey, J. (2007). Job perceptions and intent to leave among direct care workers: evidence from the better jobs better care demonstration. The Gerontologist, 47, 820- 829. Brief, A.P., Butcher, A.H., George, J.M., Link, K.E. (1993). Integrating bottom-up and top-down theories of subjective well-being: The case of health. Journal of Personality and Social Psychology, 64, 646–653. Buckley, P., Stangl, P., & Guinn, J. (2017). Why people walk: modeling foundational and higher order needs based on latent structure. Journal of Urbanism: International Research on Placemaking and Urban Sustainability, 10(2), 129- 149. Burton, E., Mitchell, L., & Stride, C. (2011). Good places for ageing in place: development of objective built environment measures for investigating links with older people’s wellbeing. BMC Public Health, 11(1), 839. Buys, L., & Miller, E. (2012). Residential satisfaction in inner urban higher-density brisbane, australia: Role of dwelling design, neighbourhood and neighbours. Journal of Environmental Planning and Management, 55(3), 319–338. http://doi.org/10.1080/09640568.2011.597592. Campbell, A., Converse, P. E., & Rodgers, W. L. (1976). The Quality of American Life: Perceptions, Evaluations, and Satisfactions. New York: Russell Sage Foundation. Cao, X. J. (2013). The association between light rail transit and satisfactions with travel and life: Evidence from twin cities. Transportation, 40(5), 921-933. Cao, X. J. (2015). Examining the impacts of neighborhood design and residential self- selection on active travel: a methodological assessment. Urban Geography, 36(2), 236-255. Cao, X. J. (2016). How does neighborhood design affect life satisfaction? Evidence from twin cities. Travel Behaviour and Society, 5, 68-76. Cao, X. J., & Ettema, D. (2014). Satisfaction with travel and residential self-selection: How do preferences moderate the impact of the Hiawatha light rail transit line?. Journal of Transport and Land Use, 7(3), 93-108.

177

References

Carlquist, E., Ulleberg, P., Delle Fave, A., Nafstad, H. E., & Blakar, R. M. (2017). Everyday understandings of happiness, good life, and satisfaction: three different facets of well-being. Applied Research in Quality of Life, 12(2), 481- 505. Carvajal, M. J., Popovici, I., & Hardigan, P. C. (2018). Gender differences in the measurement of pharmacists’ job satisfaction. Human Resources for Health, 16(1), 33. Cats, O., & Gkioulou, Z. (2017). Modeling the impacts of public transport reliability and travel information on passengers’ waiting-time uncertainty. EURO Journal on Transportation and Logistics, 6(3), 247-270. Cervero, R. (2002). Built environments and mode choice: toward a normative framework. Transportation Research Part D, Transport and Environment, 7(4), 265-284. Chang, P. J., Wray, L., & Lin, Y. (2014). Social relationships, leisure activity, and health in older adults. Health Psychology, 33(6), 516. Chapman, D. W., & Lombard, J. R. (2006). Determinants of neighborhood satisfaction in fee-based gated and non-gated communities. Urban Affairs Review, 41(6), 769–799. https://doi.org/10.1177/1078087406287164. Chaskin, R. J. (1995). Defining neighbourhood: History, theory, and practice. Chicago: Chapin hall centre for children at the university of Chicago. Chaudhuri, K., Reilly, K. T., & Spencer, D. A. (2015). Job satisfaction, age and tenure: A generalized dynamic random effects model. Economics Letters, 130, 13-16. Chen, C. F. (2008). Investigating structural relationships between service quality, perceived value, satisfaction, and behavioral intentions for air passengers: Evidence from Taiwan. Transportation Research Part A, Policy and Practice, 42(4), 709-717. Chen, L. H., Ye, Y. C., Chen, M. Y., & Tung, I. W. (2010). Alegría! Flow in leisure and life satisfaction: The mediating role of event satisfaction using data from an acrobatics show. Social Indicators Research, 99(2), 301-313. Chen, L., Zhang, W., Yang, Y., & Yu, J. (2013). Disparities in residential environment and satisfaction among urban residents in Dalian, China. Habitat International, 40, 100-108. Chevalier, A., & Feinstein, L. (2006). Sheepskin or prozac: The causal effect of education on mental health. IZA Discussion Paper, No. 2231. https://ssrn.com/abstract=923530. Churchill Jr, G. A., Ford, N. M., & Walker Jr, O. C. (1974). Measuring the job satisfaction of industrial salesmen. Journal of Marketing Research, 254-260. Clark, A. E. (1996). Job satisfaction in Britain. British Journal of Industrial Relations, 34(2), 189–217. Clark, A. E. (1999). Are wages habit-forming? Evidence from micro data. Journal of Economic Behavior & Organization, 39(2), 179-200. Clark, A. E. (2005). Your money or your life: Changing job quality in OECD countries. British Journal of Industrial Relations, 43(3), 377-400.

178

References

Clark, A. E. (2015). What makes a good job? Job quality and job satisfaction. IZA World of Labor 2015. https://doi.org/10.15185/izawol.215. Clark, W. A., Duque‐Calvache, R., & Palomares‐Linares, I. (2017). Place attachment and the decision to stay in the neighbourhood. Population, Space and Place, 23(2), 1-16. Clark, W. A., & Maas, R. (2016). Spatial mobility and opportunity in Australia: Residential selection and neighbourhood connections. Urban Studies, 53(6), 1317-1331. Cohen, P., West, S. G., & Aiken, L. S. (2014). Applied Multiple Regression/correlation Analysis for the Behavioral Sciences. New York: Psychology Press. Costa, G. (2008). Geografia della salute in contesti urbani. In Salute, Ambiente e Qualità della Vita nel Contesto Urbano [Health, Environment and Quality of Life in the Urban Context] (pp. 97–150). Milano: Franco Angeli. Costa-Font, J. (2013). Housing-related Well-being in Older People: The impact of environmental and financial influences. Urban Studies, 50(4), 657- 673. https://doi.org/10.1177/0042098012456247. Cozens, P. (2011). Urban planning and environmental criminology: towards a new perspective for safer cities. Planning Practice and Research, 26 (4), 481- 508. Cozens, P., Hillier, D., & Prescott, G. (2002). Defensible space, community safety, the British city and the active citizen: Penetrating the criminal mind. Crime Prevention and Community Safety, 4(4), 7-21. Cozens, P. , & Love, T. (2015). A review and current status of crime prevention through environmental design. Journal of Planning Literature, 30(4), 393- 412. https://doi.org/10.1177/0885412215595440. Cramm, J. M., & Nieboer, A. P. (2015). Social cohesion and belonging predict the well- being of community-dwelling older people. BMC Geriatrics, 15(1), 30. Crossman, A., & Abou-Zaki, B. (2003). Job satisfaction and employee performance of Lebanese banking staff. Journal of Managerial Psychology, 18(4), 368-376. Cummins, R. A. (1996). The domains of life satisfaction: an attempt to order chaos. Social Indicators Research, 38, 303- 332. Cummins, R. A., Eckersley, R., Pallant, J., Van Vugt, J., & Misajon, R. (2003). Developing a national index of subjective wellbeing: The Australian unity wellbeing index. Social Indicators Research, 64(2), 159-190. Currie, G., Richardson, T., Smyth, P., Vella-Brodrick, D., Hine, J., Lucas, K., Stanley, J., Morris, J., Kinnear, R., & Stanley, J. (2010). Investigating links between transport disadvantage, social exclusion and well-being in Melbourne- updated results. Research in Transportation Economics, 29(1), 287-295. Danielsson, C., & Bodin, L. (2009). Difference in satisfaction with office environment among employees in different office types. Journal of Architectural and Planning Research, 26(3), 241-257. http://www.jstor.org/stable/43030872. Darma, P. S., & Supriyanto, A. S. (2017). The effect of compensation on satisfaction and employee performance. Management and Economics Journal (MEC-J), 1(1), 69-77.

179

References

Dassopoulos, A., & Monnat, S. M. (2011). Do perceptions of social cohesion, social support, and social control mediate the effects of local community participation on neighborhood satisfaction?. Environment and Behavior, 43(4), 546- 565. https://doi.org/10.1177/0013916510366821. Dawal, S. Z. M., & Taha, Z. (2006). The effect of job and environmental factors on job satisfaction in automotive industries. International Journal of Occupational Safety and Ergonomics, 12(3), 267-280. De Croon, E., Sluiter, J., Kuijer, P. P., & Frings-Dresen, M. (2005). The effect of office concepts on worker health and performance: a systematic review of the literature. Ergonomics, 48(2), 119-134. Dekker, K., & Bolt, G. (2005). Social cohesion in post-war estates in the Netherlands: Differences between socioeconomic and ethnic groups. Urban Studies, 42(13), 2447-2470. Dekker, K., de Vos, S., Musterd, S., & Van Kempen, R. (2011). Residential satisfaction in housing estates in European cities: A multi-level research approach. Housing Studies, 26(04), 479-499. Delhey, J. (2014). Domain satisfaction. In: Michalos A.C. (eds), Encyclopedia of Quality of Life and Well-being Research (pp. 1679-1683). Dordrecht: Springer. Delle Fave, A., Brdar, I., Freire, T., Vella-Brodrick, D., & Wissing, M. P. (2011). The eudaimonic and hedonic components of happiness: qualitative and quantitative findings. Social Indicators Research, 100(2), 185-207. Dell’Olio, L., Ibeas, A., & Cecin, P. (2011). The quality of service desired by public transport users. Transport Policy, 18(1), 217-227. Den Besten, O. (2010). Local belonging and geographies of emotions: immigrant children’s experience of their neighbourhoods in Paris and Berlin. Childhood, 17(2), 181-195. de Oña, J., de Oña, R., Eboli, L., Forciniti, C., & Mazzulla, G. (2016). Transit passengers’ behavioural intentions: The influence of service quality and customer satisfaction. Transportmetrica A, Transport Science, 12(5), 385-412. De Vos, J., Mokhtarian, P. L., Schwanen, T., Van Acker, V., & Witlox, F. (2016). Travel mode choice and travel satisfaction: Bridging the gap between decision utility and experienced utility. Transportation, 43(5), 771-796. Diaz-Serrano, L. (2009). Disentangling the housing satisfaction puzzle: does homeownership really matter?. Journal of Economic Psychology, 30(5), 745- 755. http://doi.org/10.1016/j.joep.2009.06.006. Diaz-Serrano, L., & Stoyanova, A. P. (2010). Mobility and housing satisfaction: an empirical analysis for 12 EU countries. Journal of Economic Geography, 10(5), 661- 683. http://doi.org/10.1093/jeg/lbp045. Diener, E. (1984). Subjective well-being. Psychological Bulletin, 95, 542-575. Diener, E. (1994). Assessing subjective well-being: Progress and opportunities. Social Indicators Research, 31(2), 103-157. Diener, E. (2009). Subjective well-being. In E. Diener (Ed.), The science of well-being: The collected works of Ed Diener (pp. 11-58). The Netherlands: Springer.

180

References

Diener, E., Inglehart, R., & Tay, L. (2013). Theory and validity of life satisfaction scales. Social Indicators Research, 112(3), 497-527. Diener, E., & Seligman, M. E. (2002). Very happy people. Psychological Science, 13(1), 81-84. Diener, E., & Seligman, M. E. (2004). Beyond money: Toward an economy of well- being. Psychological Science in the Public Interest, 5(1), 1-31. Di Masso, A., Dixon, J., & Hernández, B. (2017). Place attachment, sense of belonging and the micro-politics of place satisfaction. In Handbook of environmental psychology and quality of life research (pp. 85-104). Cham: Springer. Ding, C., Wang, D., Liu, C., Zhang, Y., & Yang, J. (2017). Exploring the influence of built environment on travel mode choice considering the mediating effects of car ownership and travel distance. Transportation Research Part A: Policy and Practice, 100, 65-80. Djukic, M., Kovner, C. T., Brewer, C. S., Fatehi, F., & Greene, W. H. (2014). Exploring direct and indirect influences of physical work environment on job satisfaction for early-career registered nurses employed in hospitals. Research in Nursing & Health, 37(4), 312-325. Dolan, P., Peasgood, T., & White, M. (2008). Do we really know what makes us happy? A review of the economic literature on the factors associated with subjective well-being. Journal of Economic Psychology, 29(1), 94-122. Dolan, P., & White, M. (2006). Dynamic well-being: Connecting indicators of what people anticipate with indicators of what they experience. Social Indicators Research, 75(2), 303-333. Dole, C., & Schroeder, R. G. (2001). The impact of various factors on the personality, job satisfaction and turnover intentions of professional accountants. Managerial Auditing Journal, 16(4), 234-245. Dolnicar, S., Yanamandram, V., & Cliff, K. (2012). The contribution of vacations to quality of life. Annals of Tourism Research, 39(1), 59-83. Duran, M. A. (2015). The contribution of unpaid work to global wellbeing. In Global Handbook of Quality of Life (pp. 381-411). Dordrecht: Springer. Dusansky, R., & Koç, Ç. (2007). The capital gains effect in the demand for housing. Journal of Urban Economics, 61(2), 287-298. Easterlin, R. A., Angelescu, L., & Zweig, J. S. (2011). The impact of modern economic growth on urban–rural differences in subjective well-being. World Development, 39(12), 2187-2198. Easthope, H. (2004). A place called home. Housing, theory and society, 21(3), 128-138. Ellis, N., & Pompili, A. (2002). Quality of working life for nurses: report on qualitative research. Commonwealth department of health and ageing. Elsinga, M., & Hoekstra, J. (2005). Homeownership and housing satisfaction. Journal of Housing and the Built Environment, 20(4), 401- 424. Erdogan, B., Bauer, T. N., Truxillo, D. M., & Mansfield, L. R. (2012). Whistle while you work: A review of the life satisfaction literature. Journal of Management, 38(4), 1038-1083.

181

References

Ettema, D. F., Abenoza, R. F., & Susilo, Y. O. (2016). Satisfaction with intermodal trips in Stockholm: How do service attributes influence satisfaction with the main mode and with the journey as a whole?. In Proceedings of the 95th Annual Meeting of the Transportation Research Board. January 2016, Washington D.C. Ettema, D. F., Friman, M., Gärling, T., Olsson, L. E., & Fujii, S. (2012). How in-vehicle activities affect work commuters’ satisfaction with public transport?. Journal of Transport Geography, 24, 215-222. Ettema, D. F., Gärling, T., Eriksson, L., Friman, M., Olsson, L. E., & Fujii, S. (2011). Satisfaction with travel and subjective well-being: development and test of a measurement tool. Transportation Research Part F: Traffic Psychology and Behaviour, 14(3), 167-175. Ettema, D. F., Gärling, T., Olsson, L. E., & Friman, M. (2010). Out-of-home activities, daily travel, and subjective well-being. Transportation Research Part A: Policy and Practice, 44(9), 723-732. Ettema, D., & Schekkerman, M. (2016). How do spatial characteristics influence well- being and mental health? Comparing the effect of objective and subjective characteristics at different spatial scales. Travel Behaviour and Society, 5, 56- 67. Ewing, R., & Cervero, R. (2010). Travel and the built environment: a meta- analysis. Journal of the American Planning Association, 76(3), 265-294. Fagan, C. (2001). The temporal reorganization of employment and the household rhythm of work schedules: The implications for gender and class relations. American Behavioral Scientist, 44(7), 1199-1212. Fellesson, M., & Friman, M. (2008). Perceived satisfaction with public transport service in nine European cities. Journal of the Transportation Research Forum, 47(3), 93-103. Félonneau, M. L., & Causse, E. (2017). Pro-environmentalism, identity dynamics and environmental quality of life. In Handbook of environmental psychology and quality of life research, (pp. 211-228). Cham: Springer. Fernández-Carro, C., Módenes, J. A., & Spijker, J. (2015). Living conditions as predictor of elderly residential satisfaction. A cross-European view by poverty status. European Journal of Ageing, 12(3), 187-202. Fernández-Portero, C., Alarcón, D., & Padura, Á. B. (2017). Dwelling conditions and life satisfaction of older people through residential satisfaction. Journal of Environmental Psychology, 49, 1-7. Ferrer-i-Carbonell, A. (2005). Income and well-being: an empirical analysis of the comparison income effect. Journal of Public Economics, 89(5-6), 997-1019. Flavin, M., & Yamashita, T. (2002). Owner-occupied housing and the composition of the household portfolio. American Economic Review, 92(1), 345-362. Fonner, K. L., & Roloff, M. E. (2010). Why teleworkers are more satisfied with their jobs than are office-based workers: when less contact is beneficial. Journal of Applied Communication Research, 38(4), 336-361.

182

References

Fornara, F., Bonaiuto, M., & Bonnes, M. (2006). Perceived hospital environment quality indicators: A study of orthopaedic units. Journal of Environmental Psychology, 26(4), 321-334. Foster, K. A., & Hipp, J. A. (2011). Defining neighborhood boundaries for social measurement: advancing social work research. Social Work Research, 35(1), 25-35. Foye, C. (2017). The relationship between size of living space and subjective well- being. Journal of Happiness Studies, 18(2), 427-461. Frey, B. S., & Stutzer, A. (2005). Happiness research: state and prospects. Review of Social Economy, 63(2), 207-228. Friman, M., & Fellesson, M. (2009). Service supply and customer satisfaction in public transportation: The quality paradox. Journal of Public Transportation, 12(4), 4- 17. Friman, M., Fujii, S., Ettema, D. F., Gärling, T., & Olsson, L. E. (2013). Psychometric analysis of the satisfaction with travel scale. Transportation Research Part A: Policy and Practice, 48, 132-145. Friman, M., Gärling, T., Ettema, D. F., & Olsson, L. E. (2017). How does travel affect emotional well-being and life satisfaction?. Transportation research Part A: Policy and Practice, 106, 170-180. Fryers, T. (2006). Work, identity and health. Clinical Practice and Epidemiology in Mental Health, 2(1), 12-19. Furnham, A., Monsen, J., & Ahmetoglu, G. (2009). Typical intellectual engagement, Big Five personality traits, approaches to learning and cognitive ability predictors of academic performance. British Journal of Educational Psychology, 79(4), 769-782. Galster, G. (2002). An economic efficiency analysis of deconcentrating poverty populations. Journal of Housing Economics, 11(4), 303-329. Galster, G., & Hesser, G. (1981). Residential satisfaction: compositional and contextual correlates. Environment and Behavior, 13(6), 735- 758. Gao, Y. (2018). Travel Satisfaction and Subjective Well-being: a Behavioral Modeling Perspective. Eindhoven: Technische Universiteit Eindhoven. Gao, Y., Rasouli, S., Timmermans, H., & Wang, Y. (2016). Mode-switching intentions of public transport users: The influence of perceived service quality, travel satisfaction and psychological disposition. In A. W. G. Wong, S. H. F. Wong, & G. L. M. Leung (editors), Proceedings of the 21st International Conference of Hong Kong Society for Transportation Studies, HKSTS 2016 - Smart Transportation (blz. 78-85). Hong Kong Society for Transportation Studies. Gao, Y., Rasouli, S., Timmermans, H. J. P., & Wang, Y. (2017). Understanding the relationship between travel satisfaction and subjective well-being considering the role of personality traits: A structural equation model. Transportation research part F, Traffic Psychology and Behaviour, 49, 110-123. Gardner, P. J. (2011). Natural neighborhood networks- Important social networks in the lives of older adults aging in place. Journal of Aging Studies, 25(3), 263-271.

183

References

Gatersleben, B., & Uzzell, D. (2007). Affective appraisals of the daily commute comparing perceptions of drivers, cyclists, walkers, and users of public transport. Environment and Behavior, 39(3), 416-431. Gazioglu, S., & Tansel, A. (2006). Job satisfaction in Britain: individual and job related factors. Applied Economics, 38(10), 1163-1171. Gerdtham, U. G., & Johannesson, M. (2001). The relationship between happiness, health, and socio-economic factors: results based on Swedish microdata. The Journal of Socio-economics, 30(6), 553-557. Gerhart, B., & Rynes, S. (2003). Compensation: Theory, Evidence, and Strategic Implications. Thousand Oaks: SAGE publications. Gibson, J. J. (1979). The theory of affordances. In The ecological Approach to Visual Perception (pp. 127-143). New York: Psychology Press. Glatzer, W., Camfield, L., Møller, V., & Rojas, M. (2015). Global handbook of quality of life: Exploration of well-being of nations and continents. New York: Springer. Goetze, R. (2017). Private Efforts to Initiate Neighborhood Recovery. In Rebuilding America's Cities (pp. 42-51). New York: Routledge. Haar, J. M., Russo, M., Suñe, A., & Ollier-Malaterre, A. (2014). Outcomes of work–life balance on job satisfaction, life satisfaction and mental health: A study across seven cultures. Journal of Vocational Behavior, 85(3), 361-373. Hackman, R., Lawler, E., & Oldham, G. (2005). Job characteristic theory. In Miner, J. (Ed.), Organizational Behavior one: Essential theories of motivation and leadership (pp. 75-93). New York: M.E. Sharp Inc Hadavi, S., & Kaplan, R. (2016). Neighborhood satisfaction and use patterns in urban public outdoor spaces: Multidimensionality and two-way relationships. Urban Forestry & Urban Greening, 19, 110-122. Hagedorn, L. S. (2000). Conceptualizing faculty job satisfaction: components, theories, and outcomes. New Directions for Institutional Research, 2000(105), 5-20. Haggan, M. (2017). Workforce stress: Unsafe, stressed and overworked. The Australian journal of Pharmacy, 98(1167), 26. Hahn, E. A., Cella, D., Bode, R. K., & Hanrahan, R. T. (2010). Measuring social well- being in people with chronic illness. Social Indicators Research, 96(3), 381- 401. Hair Jr, J. F., Hult, G. T. M., Ringle, C., & Sarstedt, M. (2016). A primer on Partial Least Squares Structural Equation Modeling (PLS-SEM). Thousand Oaks: SAGE. Hall, M., & Greenman, E. (2013). Housing and neighborhood quality among undocumented Mexican and Central American immigrants. Social Science Research, 42(6), 1712-1725. Handy, S., & Thigpen, C. (2018). Commute quality and its implications for commute satisfaction: exploring the role of mode, location, and other factors. Travel Behaviour and Society. https://doi.org/10.1016/j.tbs.2018.03.001. Harris, K. J., & Kacmar, K. M. (2006). Too much of a good thing: the curvilinear effect of leader-member exchange on stress. The journal of Social Psychology, 146(1), 65-84.

184

References

Hauret, L., & Williams, D. R. (2017). Cross‐National analysis of gender differences in job satisfaction. Industrial relations: A journal of Economy and Society, 56(2), 203- 235. Head, A., & Lloyd-Ellis, H. (2012). Housing liquidity, mobility, and the labour market. Review of Economic Studies, 79(4), 1559- 1589. Higgins, C. D., Sweet, M. N., & Kanaroglou, P. S. (2017). All minutes are not equal: travel time and the effects of congestion on commute satisfaction in Canadian cities. Transportation, 1-20. Hincks, S., & Wong, C. (2010). The spatial interaction of housing and labour markets: commuting flow analysis of North West England. Urban Studies, 47(3), 620- 649. Hipp, J. R. (2009). Specifying the determinants of neighborhood satisfaction: A robust assessment in 24 metropolitan areas. Social Forces, 88(1), 395-424. Hipp, J. R. (2010). What is the ‘neighbourhood’ in neighbourhood satisfaction? Comparing the effects of structural characteristics measured at the micro- neighbourhood and tract levels. Urban Studies, 47(12), 2517-2536. Hipp, J. R. (2016). Collective efficacy: How is it conceptualized, how is it measured, and does it really matter for understanding perceived neighborhood crime and disorder?. Journal of Criminal Justice, 46, 32-44. Hooper, D., Coughlan, & J., Mullen, M. (2008). Structural Equation Modelling: Guidelines for Determining Model Fit. Electronic Journal of Business Research Methods, 6(1), 53-60. Hoppock, R. (1935). Job Satisfaction. Oxford, England: Harper. Hosseini Hashemzadeh, D. (2010). Investigation of the factors affecting the employee job satisfaction in industry and mine bank. Human Resource Journal, 2(5). Huang, Z., Du, X., & Yu, X. (2015). Home ownership and residential satisfaction: Evidence from Hangzhou, China. Habitat International, 49, 74-83. Hulin, C. L. (2002). Lessons from industrial and organizational psychology. In The psychology of work (pp. 19-38). New York: Psychology Press. Hur, M., & Morrow-Jones, H. (2008). Factors that influence residents' satisfaction with neighborhoods. Environment and Behavior, 40(5), 619-635. Hur, M., & Nasar, J. L. (2014). Physical upkeep, perceived upkeep, fear of crime and neighborhood satisfaction. Journal of Environmental Psychology, 38, 186-194. Hur, M., Nasar, J. L., & Chun, B. (2010). Neighborhood satisfaction, physical and perceived naturalness and openness. Journal of Environmental Psychology, 30(1), 52-59. Ibem, E. O., & Aduwo, E. B. (2013). Assessment of residential satisfaction in public housing in Ogun state, Nigeria. Habitat International, 40, 163-175. Ibem, E. O., & Amole, D. (2013). Residential satisfaction in public core housing in Abeokuta, Ogun State, Nigeria. Social Indicators Research, 113(1), 563-581. Iseki, H., & Smart, M. (2012). How do people perceive service attributes at transit facilities? Examination of perceptions of transit service by transit user demographics and trip characteristics. Transportation Research Record: Journal of the Transportation Research Board, 164-174. 185

References

Ismail, A., Bujang, A. A., Jaafar, M. N., Jiram, W. R. A., Zarin, H. A., & Maimun, N. H. A. (2018). Household’s housing affordability stress in iskandar malaysia: a review. Advanced Science Letters, 24(6), 4295-4298. Jacobs, J. (1961). The Death and Life of Great American Cities. New-York: Vintage. Jain, R., & Kaur, S. (2014). Impact of work environment on job satisfaction. International Journal of Scientific and Research Publications, 4(1), 1-8. Jensen, W. A., Stump, T. K., Brown, B. B., Werner, C. M., & Smith, K. R. (2017). Walkability, complete streets, and gender: who benefits most?. Health & Place, 48, 80-89. Jiang, W., Feng, T., Timmermans, H., & Li, H. (2017). A gap-theoretical path model of residential satisfaction and intention to move house applied to renovated historical blocks in two Chinese cities. Cities, 71(11), 19–29. Jiang, Z., & Hu, X. (2016). Knowledge sharing and life satisfaction: The roles of colleague relationships and gender. Social Indicators Research, 126(1), 379- 394. Jiboye, A. D. (2014). Significance of house-type as a determinant of residential quality in Osogbo, Southwest Nigeria. Frontiers of Architectural Research, 3(1), 20-27. Joewono, T. B., & Kubota, H. (2006). Safety and security improvement in public transportation based on public perception in developing countries. IATSS Research, 30(1), 86-100. Judge, T. A., Piccolo, R. F., Podsakoff, N. P., Shaw, J. C., & Rich, B. L. (2010). The relationship between pay and job satisfaction: A meta-analysis of the literature. Journal of Vocational Behavior, 77(2), 157-167. Jussila, I., Tarkiainen, A., Sarstedt, M., & Hair, J. F. (2015). Individual psychological ownership: Concepts, evidence, and implications for research in marketing. Journal of Marketing Theory and Practice, 23(2), 121-139. Kahneman, D., & Deaton, A. (2010). High income improves evaluation of life but not emotional well-being. Proceedings of the National Academy of Sciences, 107(38), 16489-16493. Kahneman, D., Diener, E., & Schwarz, N. (1999). Well-Being: foundations of hedonic psychology. New York: Russell SAGE Foundation. Kahneman, D., Krueger, A. B., Schkade, D. A., Schwarz, N., & Stone, A. A. (2004). A survey method for characterizing daily life experience: the day reconstruction method. Science, 306, 1776–1780. Karatuna, I., & Basol, O. (2017). Job satisfaction of part-time vs. full-time workers in Turkey: the effect of income and work status satisfaction. International Journal of Value Chain Management, 8(1), 58-72. Kauhanen, M., & Nätti, J. (2015). Involuntary temporary and part-time work, job quality and well-being at work. Social Indicators Research, 120(3), 783-799. Keul, A. G., & Prinz, T. (2011). The Salzburg quality of urban life study with GIS support. In Investigating Quality of Urban Life, (pp. 273-293). Dordrecht: Springer.

186

References

Kline, R. B. (1998). Software review: software programs for structural equation modeling: Amos, EQS, and LISREL. Journal of Psychoeducational Assessment, 16(4), 343-364. Koohsari, M. J., Badland, H., Sugiyama, T., Mavoa, S., Christian, H., & Giles-Corti, B. (2015). Mismatch between perceived and objectively measured land use mix and street connectivity: associations with neighborhood walking. Journal of Urban Health, 92(2), 242-252. Kweon, B. S., Ellis, C. D., Leiva, P. I., & Rogers, G. O. (2010). Landscape components, land use, and neighborhood satisfaction. Environment and Planning B: Planning and Design, 37(3), 500-517. Lance, C. E., Mallard, A. G. C., & Michalos, A. C. (1995). Tests of causal directions of global-life facet satisfaction relationships. Social Indicators Research, 34 , 69– 92. Lane, R. E. (1994). Quality of life and quality of persons: a new role for government?. Political Theory, 22(2), 219-252. Lee, J. S., Back, K. J., & Chan, E. S. (2015). Quality of work life and job satisfaction among frontline hotel employees: A self-determination and need satisfaction theory approach. International journal of Contemporary Hospitality Management, 27(5), 768-789. Lee, S. M., Conway, T. L., Frank, L. D., Saelens, B. E., Cain, K. L., & Sallis, J. F. (2017). The relation of perceived and objective environment attributes to neighborhood satisfaction. Environment and Behavior, 49(2), 136-160. Lee, S. Y., & Brand, J. L. (2010). Can personal control over the physical environment ease distractions in office workplaces?. Ergonomics, 53(3), 324-335. Lee, T. (1968). Urban neighbourhood as a socio-spatial schema. Human Relations, 21(3), 241-267. Legrain, A., Eluru, N., & El-Geneidy, A. M. (2015). Am stressed, must travel: the relationship between mode choice and commuting stress. Transportation Research Part F: Traffic Psychology and Behaviour, 34, 141-151. Levy, P. E., Cober, R. T., & Norris-Watts, C. (2004). Specific personality measures. In Comprehensive Handbook of Psychological Assessment (pp. 130-148). Hoboken: John Wiley & Sons. Li, S., & Song, Y. (2009). Redevelopment, Displacement, Housing Conditions, and Residential Satisfaction: A study of Shanghai. Environment and Planning A, 41(5), 1090–1108. http://doi.org/10.1068/a4168. Li, Z., & Wu, F. (2013). Residential satisfaction in China’s informal settlements: a case study of Beijing, Shanghai, and Guangzhou. Urban Geography, 34(7), 923– 949. http://doi.org/10.1080/02723638.2013.778694. Locke, E. A. (1968). Toward a theory of task motivation and incentives. Organizational Behavior and Human Performance, 3(2), 157-189. Loewe, N., Bagherzadeh, M., Araya-Castillo, L., Thieme, C., & Batista-Foguet, J. M. (2014). Life domain satisfactions as predictors of overall life satisfaction among workers: Evidence from Chile. Social Indicators Research, 118(1), 71-86.

187

References

Lottrup, L., Stigsdotter, U. K., Meilby, H., & Claudi, A. G. (2015). The workplace window view: a determinant of office workers’ work ability and job satisfaction. Landscape Research, 40(1), 57-75. Lovejoy, K., Handy, S., & Mokhtarian, P. (2010). Neighborhood satisfaction in suburban versus traditional environments: An evaluation of contributing characteristics in eight California neighborhoods. Landscape and Urban Planning, 97(1), 37-48. Low, C. T., Stimson, R., Chen, S., Cerin, E., Wong, P. P. Y., & Lai, P. C. (2018). Personal and neighbourhood indicators of quality of urban life: a case study of Hong Kong. Social Indicators Research, 136(2), 751-773. Lu, M. (1999). Determinants of residential satisfaction: Ordered logit vs. regression models. Growth and Change, 30(2), 264-287. http://doi.org/10.1111/0017- 4815.00113. Lu, M. (2002). Are Pastures Greener? Residential consequences of migration. International Journal of Population Geography, 8(3), 201–216. http://doi.org/10.1002/ijpg.244. Lynch, K. (1960). The Image of the City. Massachusetts: MIT press. Lynch, K. (1984). Good City Form. Massachusetts: MIT press. Maat, K., & Timmermans, H. J. (2009). A causal model relating urban form with daily travel distance through activity/travel decisions. Transportation Planning and Technology, 32(2), 115-134. Macke, J., Casagrande, R. M., Sarate, J. A. R., & Silva, K. A. (2018). Smart city and quality of life: Citizens’ perception in a Brazilian case study. Journal of Cleaner Production, 182, 717-726. Makinde, O. O. (2015). Influences of socio-cultural experiences on residents’ satisfaction in Ikorodu low-cost housing estate, Lagos state. Environment, Development and Sustainability, 17(1), 173-198. Marans, R. W. (2012). Quality of urban life studies: An overview and implications for environment-behaviour research. Procedia-Social and Behavioral Sciences, 35, 9-22. Marans, R. W. (2015). Quality of urban life & environmental sustainability studies: future linkage opportunities. Habitat International, 45, 47-52. Marans, R. W., & Kweon, B. S. (2011). The quality of life in metro Detroit at the beginning of the millennium. In Investigating Quality of Urban Life (pp. 163- 183). Dordrecht: Springer. Marans, R. W., & Rodgers, W. (1975). Toward an Understanding of Community Satisfaction. In Metropolitan America in Contemporary Perspective (pp. 299- 352). New York: Halsted Press. Marans, R. W., & Stimson, R. J. (Eds.). (2011). Investigating Quality of Urban Life: Theory, Methods, and Empirical Research (Vol. 45). London: Springer Science & Business Media. Marcouyeux, A., & Fleury-Bahi, G. (2011). Place-identity in a school setting: effects of the place image. Environment and Behavior, 43(3), 344-362. Maslow, A. H. (1954). Motivation and Personality. New York: Harper and Row.

188

References

Mas-Machuca, M., Berbegal-Mirabent, J., & Alegre, I. (2016). Work-life balance and its relationship with organizational pride and job satisfaction. Journal of Managerial Psychology, 31(2), 586-602. Massey, D. B. (1995). The Conceptualization of Place. In A Place in the World? Places, Cultures and Globalization (pp. 45-85). Oxford, New York: Oxford University Press. Mattisson, K., Håkansson, C., & Jakobsson, K. (2015). Relationships between commuting and social capital among men and women in southern Sweden. Environment and Behavior, 47(7), 734-753. Mattisson, K., Jakobsson, K., Håkansson, C., & Cromley, E. (2016). Spatial heterogeneity in repeated measures of perceived stress among car commuters in Scania, Sweden. International Journal of Health Geographics, 15(1), 22. McCrea, R., Stimson, R., & Western, J. (2005). Testing a moderated model of satisfaction with urban living using data for Brisbane-South East Queensland, Australia. Social Indicators Research, 72(2), 121-152. McCrea, R., Western, M., & Shyy, T. K. (2011). Subjective Quality of Life in Queensland: Comparing Metropolitan, Regional and Rural Areas. In Investigating Quality of Urban Life (pp. 295-313). Dordrecht: Springer. Meegan, R., & Mitchell, A. (2001). It's not community round here, it's neighbourhood: neighbourhood change and cohesion in urban regeneration policies. Urban Studies, 38(12), 2167-2194. Mitchell, A., Macció, J., & Fages, D. M. (2018). The Effects of Emergency Housing on Wellbeing: Evidence from Argentina’s Informal Settlements. The European Journal of Development Research, 1-26. https://doi.org/10.1057/s41287-018- 0166-z. Mohit, M. A., Ibrahim, M., & Rashid, Y. R. (2010). Assessment of residential satisfaction in newly designed public low-cost housing in Kuala Lumpur, Malaysia. Habitat International, 34(1), 18-27. http://doi.org/10.1016/j.habitatint.2009.04.002. Mohit, M. A., & Nazyddah, N. (2011). Social housing programme of Selangor Zakat Board of Malaysia and housing satisfaction. Journal of Housing and the Built Environment, 26(2), 143-164. Mokhtarian, P. L., & Pendyala, R. M. (2018). Travel satisfaction and well-being. In Quality of Life and Daily Travel (pp. 17-39). Cham: Springer. Molin, E. J. E., & Timmermans, H. J. P. (2003). Testing hierarchical information integration theory: the causal structure of household residential satisfaction. Environment and Planning A, 35(1), 43–58. http://doi.org/10.1068/a3416. Montero, R., & Rau, T. (2015). Part-time work, job satisfaction and well-being: evidence from a developing OECD country. The Journal of Development Studies, 51(4), 370-385. Morris, E. A., & Guerra, E. (2015). Mood and mode: does how we travel affect how we feel?. Transportation, 42(1), 25-43. Morris, E. A., & Hirsch, J. A. (2016). Does rush hour see a rush of emotions? driver mood in conditions likely to exhibit congestion. Travel Behaviour and Society, 5, 5-13.

189

References

Moss, S. (2009). Fit indices for structural equation modeling. Website: http://www. psych-it. com. au/Psychlopedia/article. asp. Müller, T. (2017). The social construction of the stigma of an inner-city neighborhood. Conflicting perspectives of professionals and residents on social problems and gentrification. In Jahrbuch für Soziale Interaktion 1 (pp. 47-67). Wiesbaden: Springer VS. Mumford, L. (1954). The neighborhood and the neighborhood unit. Town Planning Review, 24(4), 256. Munshi, T. (2016). Built environment and mode choice relationship for commute travel in the city of Rajkot, India. Transportation Research Part D: Transport and Environment, 44, 239-253. Muthén, L. K., & Muthén, B. O. (2002). How to use a Monte Carlo study to decide on sample size and determine power. Structural Equation Modeling, 9(4), 599- 620. Nanjundeswaraswamy, T. S., & Swamy, D. R. (2013). quality of worklife of employees in private technical institutions. International Journal for Quality Research, 7(3), 431-441. Newman, O. (1972). Defensible space. New York: MacMillan. Nguyen, P. D., Dang, C. X., & Nguyen, L. D. (2015). Would better earning, work environment, and promotion opportunities increase employee performance? An investigation in state and other sectors in Vietnam. Public Organization Review, 15(4), 565-579. Novaco, R. W., & Gonzalez, O. I. (2009). Commuting and well-being. Technology and Well-being, 3, 174-205. Nyberg, A. J., Maltarich, M. A., Abdulsalam, D. D., Essman, S. M., & Cragun, O. (2018). Collective pay for performance: a cross-disciplinary review and meta- analysis. Journal of Management, 44(6), 2433-2472. Oakil, A. T. M., Nijland, L., & Dijst, M. (2016). Rush hour commuting in the Netherlands: Gender-specific household activities and personal attitudes towards responsibility sharing. Travel Behaviour and Society, 4, 79-87. Ogu, V. I. (2002). Urban residential satisfaction and the planning implications in a developing world context: The example of Benin City, Nigeria. International Planning Studies, 7(1), 37-53. Oh, J. H. (2003). Social bonds and the migration intentions of elderly urban residents: The mediating effect of residential satisfaction. Population Research and Policy Review, 22(2), 127-146. Oktay, D., & Rustemli, A. (2011). The quality of urban life and neighborhood satisfaction in Famagusta, Northern Cyprus. In Investigating Quality of Urban Life (pp. 233-249). Dordrecht: Springer. Okulicz-Kozaryn, A. (2013). City life: rankings (livability) versus perceptions (satisfaction). Social Indicators Research, 110(2), 433-451. Okulicz-Kozaryn, A., & Mazelis, J. M. (2018). Urbanism and happiness: a test of Wirth’s theory of urban life. Urban Studies, 55(2), 349-364.

190

References

Olsen, J. R., Macdonald, L., & Ellaway, A. (2017). Changes over time in population level transport satisfaction and mode of travel: a 13 year repeat cross-sectional study, UK. Journal of Transport & Health, 6, 366-378. Olsson, L. E., Gärling, T., Ettema, D. F., Friman, M., & Fujii, S. (2013). Happiness and satisfaction with work commute. Social Indicators Research, 111(1), 255-263. Orstad, S. L., McDonough, M. H., Stapleton, S., Altincekic, C., & Troped, P. J. (2017). A systematic review of agreement between perceived and objective neighborhood environment measures and associations with physical activity outcomes. Environment and Behavior, 49(8), 904-932. Orth, U., Trzesniewski, K. H., & Robins, R. W. (2010). Self-esteem development from young adulthood to old age: a cohort-sequential longitudinal study. Journal of Personality and Social Psychology, 98(4), 645. Ory, D. T., & Mokhtarian, P. L. (2005). When is getting there half the fun? Modeling the liking for travel. Transportation Research Part A, Policy and Practice, 39(2), 97- 123. Oud, J. H., & Folmer, H. (2008). A structural equation approach to models with spatial dependence. Geographical Analysis, 40(2), 152-166. Pacione, M. (2011). The geography of the Third World: progress and prospect. London: Routledge. Parkes, A., Kearns, A., & Atkinson, R. (2002). What makes people dissatisfied with their neighbourhoods?. Urban Studies, 39(13), 2413-2438. Peiser, R. B. (2016). From spontaneous to planned urban development and quality of life: the case of Ho Chi Minh City. Applied Research in Quality of Life, 11(4), 1357-1377. Pekkonen, M., & Haverinen-Shaughnessy, U. (2015). Housing satisfaction in Finland with regard to area, dwelling type and tenure status. Central European Journal of Public Health, 23(4), 314. Pietersma, S., de Vries, M., & van den Akker-van, M. E. (2014). Domains of quality of life: results of a three-stage Delphi consensus procedure among patients, family of patients, clinicians, scientists and the general public. Quality of Life Research, 23(5), 1543-1556. Pinto, S., Fumincelli, L., Mazzo, A., Caldeira, S., & Martins, J. C. (2017). Comfort, well- being and quality of life: discussion of the differences and similarities among the concepts. Porto Biomedical Journal, 2(1), 6-12. Pocock, B. (2005). Work‐life balance in Australia: limited progress, dim prospects. Asia Pacific Journal of Human Resources, 43(2), 198-209. Proshansky, H. M, Ittelson, W. H., & Rivlin, L. G. (1970). Environmental psychology: Man and his physical setting. Oxford, England: Holt, Rinehart & Winston. Qureshi, M. I., Iftikhar, M., Abbas, S. G., Hassan, U., Khan, K., & Zaman, K. (2013). Relationship between job stress, workload, environment and employees turnover intentions: What we know, what should we know. World Applied Sciences Journal, 23(6), 764-770.

191

References

Raymond, C. M., Kyttä, M., & Stedman, R. (2017). Sense of place, fast and slow: the potential contributions of affordance theory to sense of place. Frontiers in Psychology, 8, 1674. Redman, L., Friman, M., Gärling, T., & Hartig, T. (2013). Quality attributes of public transport that attract car users: a research review. Transport Policy, 25, 119- 127. Rezvani, M. R., Mansourian, H., & Sattari, M. H. (2013). Evaluating quality of life in urban areas (case study: Noorabad City, Iran). Social indicators research, 112(1), 203-220. Rice, R. W. (1984). Organizational Work and the Overall Quality of Life. State University of New York at Buffalo Department of Psychology. https://apps.dtic.mil/dtic/tr/fulltext/u2/a145596.pdf. Rioux, L. (2017). Comfort at Work: an indicator of quality of life at work. In Handbook of Environmental Psychology and Quality of Life Research (pp. 401-419). Cham: Springer. Rioux, L., & Werner, C. (2011). Residential satisfaction among aging people living in place. Journal of Environmental Psychology, 31(2), 158-169. Roberts-Hughes, R. (2011). The Case for Space: the Size of England’s New Homes. London: Royal Institute of British Architects. Rollings, K. A., Wells, N. M., Evans, G. W., Bednarz, A., & Yang, Y. (2017). Housing and neighborhood physical quality: Children's mental health and motivation. Journal of Environmental Psychology, 50, 17-23. Rosenberg, M. J., Hovland, C. I., McGuire, W. J., Abelson, R. P., & Brehm, J. W. (1960). Attitude organization and change: An analysis of consistency among attitude components. Yale’s studies in attitude and communication, 3. Russo, M., & Buonocore, F. (2012). The relationship between work-family enrichment and nurse turnover. Journal of Managerial Psychology, 27(3), 216-236. Ryan, R. M., & Deci, E. L. (2001). On happiness and human potentials: a review of research on hedonic and eudaimonic well-being. Annual Review of Psychology, 52(1), 141-166. Saeed, I., Waseem, M., Sikander, S., & Rizwan, M. (2014). The relationship of turnover intention with job satisfaction, job performance, leader member exchange, emotional intelligence and organizational commitment. International Journal of Learning and Development, 4(2), 242-256. Safi, M. (2009). Immigrants’ life satisfaction in Europe: between assimilation and discrimination. European Sociological Review, 26(2), 159-176. Sahin, H., & Sahingoz, S. A. (2013). The relationship between personnel workload and work satisfaction within accommodation and nutrition establishments. American International Journal of Social Science, 2(3), 10-18. Salleh, N. A., Yusof, N. A., Salleh, A. G., & Johari, D. N. (2011). Tenant satisfaction in public housing and its relationship with rent arrears: Majlis Bandaraya Ipoh, Perak, Malaysia. International Journal of Trade, Economics and Finance, 2(1), 10-18.

192

References

Santos, L. D., Martins, I. & Brito, P. (2007): Measuring subjective quality of life: A survey to Porto’s residents. Applied Research in Quality of Life, 2, 51–64. Saraji, G. N., & Dargahi, H. (2006). Study of quality of work life. Iranian Journal of Public Health, 35(4), 8-14. Schalock, R. L., Keith, K. D., Hoffman, K., & Karan, O. C. (1989). Quality of life: its measurement and use. Mental Retardation, 27(1), 25. Schepers, P., Fishman, E., Beelen, R., Heinen, E., Wijnen, W., & Parkin, J. (2015). The mortality impact of bicycle paths and lanes related to physical activity, air pollution exposure and road safety. Journal of Transport & Health, 2(4), 460- 473. Schimmack, U. (2008). The structure of subjective well-being. In The Science of Subjective Well-being (pp. 97–12). New York: Guilford Press. Schwanen, T., Kwan, M. P., & Ren, F. (2008). How fixed is fixed? Gendered rigidity of space–time constraints and geographies of everyday activities. Geoforum, 39(6), 2109-2121. Sealy–Jefferson, S., Messer, L., Slaughter-Acey, J., & Misra, D. P. (2017). Inter- relationships between objective and subjective measures of the residential environment among urban African American women. Annals of Epidemiology, 27(3), 164-168. Sedaghatnia, S., Lamit, H., Ghahramanpouri, A., & Mohamad, S. B. (2013). An evaluation of residents’ quality of life through neighborhood satisfaction in Malaysia. Environmental Management and Sustainable Development, 2(1), 114-125. Selvarajan, T. T., Slattery, J., & Stringer, D. Y. (2015). Relationship between gender and work related attitudes: a study of temporary agency employees. Journal of Business Research, 68(9), 1919-1927. Sen, A. (1993). Capability and Weil-Being. In The Quality of Life (pp.35-53). Oxford: Clarendon Press. Senasu, K., & Singhapakdi, A. (2017). Determinants of happiness in Thailand: the moderating role of religiousness. Journal of Human Behavior in the Social Environment, 27(4), 270-290. Senlier, N., Yildiz, R., & Aktaş, E. D. (2009). A perception survey for the evaluation of urban quality of life in Kocaeli and a comparison of the life satisfaction with the European cities. Social Indicators Research, 94(2), 213-226. Şimsek, Ö. F. (2009). Happiness revisited: Ontological well-being as a theory-based construct of subjective well-being. Journal of Happiness Studies, 10(5), 505- 522. Sirgy, M. J. (2012). The psychology of quality of life: Hedonic well-being, life satisfaction, and eudaimonia (Vol. 4). Dordrecht: Springer Science & Business Media. Sirgy M.J., & Cornwell, T. (2002). How neighborhood features affect quality of life. Social Indicators Research, 59(1), 79–114. Sirgy, M. J., Kruger, P. S., Lee, D. J., & Yu, G. B. (2011). How does a travel trip affect tourists’ life satisfaction?. Journal of Travel Research, 50(3), 261-275.

193

References

Sirgy, M. J., & Lee, D. J. (2016). Work-life balance: A quality-of-life model. Applied Research in Quality of Life, 11(4), 1059-1082. Sirgy, M. J., Rahtz, D. R., Cicic, M., & Underwood, R. (2000). A method for assessing residents' satisfaction with community-based services: a quality-of-life perspective. Social Indicators Research, 49(3), 279-316. Sirgy, M. J., Widgery, R. N., Lee, D. J., & Grace, B. Y. (2010). Developing a measure of community well-being based on perceptions of impact in various life domains. Social Indicators Research, 96(2), 295-311. Spears, S., Houston, D., & Boarnet, M. G. (2013). Illuminating the unseen in transit use: a framework for examining the effect of attitudes and perceptions on travel behavior. Transportation Research Part A: Policy and Practice, 58, 40-53. Spencer, E. S., Deal, A. M., Pruthi, N. R., Gonzalez, C. M., Kirby, E. W., Langston, J., ... & Wallen, E. M. (2016). Gender differences in compensation, job satisfaction and other practice patterns in urology. The Journal of Urology, 195(2), 450- 455. Stanley, J., Hensher, D. A., Stanley, J., & Vella-Brodrick, D. (2011). Mobility, social exclusion and well-being: exploring the links. Transportation Research Part A: Policy and Practice, 45(8), 789-801. Steg, L. (2005). Car use: lust and must. Instrumental, symbolic and affective motives for car use. Transportation Research Part A: Policy and Practice, 39(2-3), 147- 162. Steger, M. F. (2013). Experiencing meaning in life: Optimal functioning at the nexus of well-being, psychopathology, and spirituality. In The human quest for meaning: Theories, Research, and Applications (pp. 211-230). New York: Routledge. Steijn, B. (2004). Human resource management and job satisfaction in the Dutch public sector. Review of Public Personnel Administration, 24(4), 291-303. Stimson, R., McCrea, R., & Western, J. (2011). The Brisbane-South East Queensland region, Australia: Subjective assessment of quality of urban life and changes over time. In Investigating Quality of Urban Life (pp. 185-207). Dordrecht: Springer. St-Louis, E., Manaugh, K., van Lierop, D., & El-Geneidy, A. (2014). The happy commuter: a comparison of commuter satisfaction across modes. Transportation Research Part F: Traffic Psychology and Behaviour, 26, 160- 170. Stradling, S. G., Anable, J., & Carreno, M. (2007). Performance, importance and user disgruntlement: A six-step method for measuring satisfaction with travel modes. Transportation Research Part A: Policy and Practice, 41(1), 98-106. Stutzer, A., & Frey, B. S. (2008). Stress that doesn't pay: The commuting paradox. Scandinavian Journal of Economics, 110(2), 339-366. Sukriket, P. (2014). The relationship between job satisfaction and turnover intention of Thai software programmers in Bangkok, Thailand. AU Journal of Management, 12(2), 42-52.

194

References

Susilo, Y. O., & Cats, O. (2014). Exploring key determinants of travel satisfaction for multimodal trips by different traveler groups. Transportation Research Part A: Policy and Practice, 67, 366-380. Talen, E. (2008). New urbanism, social equity, and the challenge of post-Katrina rebuilding in Mississippi. Journal of Planning Education and Research, 27(3), 277-293. Talen, E., & Shah, S. (2007). Neighborhood evaluation using GIS: an exploratory study. Environment and Behavior, 39(5), 583-615. Tao, L., Hui, E. C., Wong, F. K., & Chen, T. (2015). Housing choices of migrant workers in China: Beyond the Hukou perspective. Habitat International, 49, 474-483. Tay, L., Kuykendall, L., & Diener, E. (2015). Satisfaction and happiness–the bright side of quality of life. In Global Handbook of Quality of Life (pp. 839-853). Dordrecht: Springer. Teck-Hong, T. (2012). Housing satisfaction in medium-and high-cost housing: the case of Greater Kuala Lumpur, Malaysia. Habitat International, 36(1), 108-116. Temelová, J., & Dvořáková, N. (2012). Residential satisfaction of elderly in the city centre: the case of revitalizing neighbourhoods in Prague. Cities, 29(5), 310– 317. http://doi.org/10.1016/j.cities.2011.11.015. Theodori, G. L. (2001). Examining the effects of community satisfaction and attachment on individual well‐being. Rural Sociology, 66(4), 618-628. Turcotte, M. (2005). The time it takes to get to work and back. Ottawa: Statistics Canada, Minister of Industry. Türkoğlu, H., Bölen, F., Baran, P. K., & Terzi, F. (2011). Measuring quality of urban life in Istanbul. In Investigating Quality of Urban Life (pp. 209-231). Dordrecht: Springer. Türksever, A. N. E., & Atalik, G. (2001). Possibilities and limitations for the measurement of the quality of life in urban areas. Social Indicators Research, 53(2), 163-187. Ullman, J. B. (2006). Structural equation modeling: Reviewing the basics and moving forward. Journal of Personality Assessment, 87(1), 35-50. Urry, J. (2016). Mobilities: New Perspectives on Transport and Society. New York: Routledge. Van Aerden, K., Moors, G., Levecque, K., & Vanroelen, C. (2015). The relationship between employment quality and work-related well-being in the European labor force. Journal of Vocational Behavior, 86, 66-76. Van Assche, J., Asbrock, F., Roets, A., & Kauff, M. (2018). Positive neighborhood norms buffer ethnic diversity effects on neighborhood dissatisfaction, perceived neighborhood disadvantage, and moving intentions. Personality and Social Psychology Bulletin, 44(5), 700-716. Van Cauwenberg, J., Clarys, P., De Bourdeaudhuij, I., Ghekiere, A., de Geus, B., Owen, N., & Deforche, B. (2018). Environmental influences on older adults’ transportation cycling experiences: a study using bike-along interviews. Landscape and Urban Planning, 169, 37-46.

195

References

Van den Berg, P. (2012). Social activity-travel patters: the role of personal networks and communication technology, PhD Dissertation. Eindhoven: Technische Universiteit Eindhoven. Van der Meer, P. H. (2014). Gender, unemployment and subjective well-being: why being unemployed is worse for men than for women. Social Indicators Research, 115(1), 23-44. Van Praag, B. M., Van Praag, B., & Ferrer-i-Carbonell, A. (2004). Happiness quantified: a satisfaction calculus approach. Oxford: Oxford University Press. Varghese, S., & Jayan, C. (2013). Quality of work life: a dynamic multidimensional construct at work place, part II. Guru Journal of Behavioral and Social Sciences, 1(2), 91-104. Veenhoven, R., & Ehrhardt, J. (1995). The cross-national pattern of happiness: test of predictions implied in three theories of happiness. Social Indicators Research, 34(1), 33-68. Veitch, J. A. (2011). Workplace design contributions to mental health and well- being. Healthcare Papers, 11, 38-46. Vera-Toscano, E., & Ateca-Amestoy, V. (2008). The relevance of social interactions on housing satisfaction. Social Indicators Research, 86(2), 257-274. Vestling, M., Tufvesson, B., & Iwarsson, S. (2003). Indicators for return to work after stroke and the importance of work for subjective well-being and life satisfaction. Journal of Rehabilitation Medicine, 35(3), 127-131. Vischer, J. C. (2007). The effects of the physical environment on job performance: towards a theoretical model of workspace stress. Stress and health: Journal of the International Society for the Investigation of Stress, 23(3), 175-184. Vischer, J. C., & Fischer, G. N. (2005). User evaluation of the work environment: a diagnostic approach. Le Travail Humain, 68(1), 73-96. Visser, K., Bolt, G., & van Kempen, R. (2015). Come and live here and you'll experience it: youths talk about their deprived neighbourhood. Journal of Youth Studies, 18(1), 36-52. Von Wirth, T., Grêt-Regamey, A., & Stauffacher, M. (2015). Mediating effects between objective and subjective indicators of urban quality of life: testing specific models for safety and access. Social Indicators Research, 122(1), 189-210. Wan, D., Kamga, C., Hao, W., Sugiura, A., & Beaton, E. B. (2016). Customer satisfaction with bus rapid transit: a study of New York City select bus service applying structural equation modeling. Public Transport, 8(3), 497-520. Wang, D., & Wang, F. (2016). Contributions of the usage and affective experience of the residential environment to residential satisfaction. Housing Studies, 31(1), 42-60. Wang, W. C., Kao, C. H., Huan, T. C., & Wu, C. C. (2011). Free time management contributes to better quality of life: a study of undergraduate students in Taiwan. Journal of Happiness Studies, 12(4), 561-573. Washington, S. P., Karlaftis, M. G., & Mannering, F. (2010). Statistical and Econometric Methods for Transportation Data Analysis. New York: Chapman and Hall/CRC.

196

References

Wasko, M. M., & Faraj, S. (2005). Why should I share? Examining social capital and knowledge contribution in electronic networks of practice. MIS Quarterly, 35- 57. Węziak-Białowolska, D. (2016). Quality of life in cities–Empirical evidence in comparative European perspective. Cities, 58, 87-96. Wittmer, J. L., & Martin, J. E. (2011). Work and personal role involvement of part‐time employees: implications for attitudes and turnover intentions. Journal of Organizational Behavior, 32(5), 767-787. Woo, E., Kim, H., & Uysal, M. (2016). A measure of quality of life in elderly tourists. Applied Research in Quality of Life, 11(1), 65-82. Wyatt, T. A., & Wah, C. Y. (2001). Perceptions of QWL: a study of Singaporean employees development. Research and Practice in Human Resource Management, 9(2), 59-76. Yavuz, N., & Welch, E. W. (2010). Addressing fear of crime in public space: gender differences in reaction to safety measures in train transit. Urban Studies, 47(12), 2491-2515. Ye, R., & Titheridge, H. (2016). Commuting satisfaction in lower income population: a case study of Xi’an, China. In Proceedings of the 95th Annual Meeting of the Transportation Research Board. January 2016, Washington D.C. Ye, R., & Titheridge, H. (2017). Satisfaction with the commute: the role of travel mode choice, built environment and attitudes. Transportation Research Part D: Transport and Environment, 52, 535-547. Zacher, H., & Griffin, B. (2015). Older workers’ age as a moderator of the relationship between career adaptability and job satisfaction. Work, Aging and Retirement, 1(2), 227-236. Zavei, S. J. A. P., & Jusan, M. M. (2012). Exploring housing attributes selection based on Maslow's hierarchy of needs. Procedia-Social and Behavioral Sciences, 42, 311-319. Zebardast, E., & Nooraie, H. (2018). Investigating the relationship between housing satisfaction and quality of life in the decayed historic areas of Isfahan using path diagram. Indoor and Built Environment, 27(5), 645-657. Zhang, C., Juan, Z., Lu, W., & Xiao, G. (2016). Do the organizational forms affect passenger satisfaction? Evidence from Chinese public transport service. Transportation Research Part A: Policy and Practice, 94, 129-148. Zhang, C., & Lu, B. (2016). Residential satisfaction in traditional and redeveloped inner city neighborhood: A tale of two neighborhoods in Beijing. Travel Behaviour and Society, 5, 23-36. Zhang, F., Zhang, C., & Hudson, J. (2018). Housing conditions and life satisfaction in urban China. Cities, 81, 35-44. Zhang, Z., & Zhang, J. (2017). Perceived residential environment of neighborhood and subjective well-being among the elderly in China: a mediating role of sense of community. Journal of Environmental Psychology, 51, 82-94. Zhu, Y. (2012). A review of job satisfaction. Asian Social Science, 9(1), 293.

197

References

198

Appendix: QOUL Survey

199

Appendix: QOUL Survey

200

Appendix: QOUL Survey

201

Appendix: QOUL Survey

Kies welke kenmerken van toepassing zijn op uw huis en waardeer deze

Huiseigen Soort Woonoppervla Aantal Buitenruimte Bouwjaar dom woonhuis kte (m2) slaapkamers

Appartement Balkon Na 2010 Huur Minder dan 60 Geen Tuin 2001-2010 Rijtjeswoning Eén Dakterras 1991-2000 Optie Koop 60-100 Twee Balkon en tuin 1981-1990 Twee-onder- Drie of meer Balkon en 1971-1980 één- kapwoning 100 en groter dakterras 1960-1970 Tuin en dakterras 1945-1959 Vrijstaande Geen 1931-1944 woning buitenruimte 1906-1930 Voor 1906

Tevredenheid (1-10) vervolgens van 1 (zeer ontevreden) tot 10 (zeer tevreden).

Hoofdweergave Binnen/buiten Maandelijkse Veiligheid van Lay-out Energiebron van het huis kwaliteit prijs van de het huis (ontwerp en de plattegrond) (reparatie woning situatie) (exclusief energie)

Groenrijke Grote omgeving reparaties of 0 Euro Mijn huis heeft Het past bij Elektriciteit renovaties zijn niet geen speciale mijn leefstijl

Blokken met nodig 500-750 beveiliging Gas en huizen Het past niet elektriciteit Het dak 750-1000 Mijn huis heeft bij mijn leefstijl Straat met traditionele Elektriciteit en Optie winkels De vloer 1000-1250 beveiliging zoals warm water speciale sloten op Snelweg De badkamer 1250-1500 deuren en ramen.. /hond Braakliggend De keuken 1500 en hoger terrain Mijn huis heeft Leidingen of high-tech beveiliging Anders bekabeling zoals camera's, alarm,... Verschillende onderdelen van het huis Tevredenheid (1-10)

Nu rekening houdend met alles, hoe tevreden bent u over uw huis? (van 1 zeer ontevreden tot 10 zeer tevreden). 1 2 3 4 5 6 7 8 9 10

202

Appendix: QOUL Survey

203

Appendix: QOUL Survey

204

Appendix: QOUL Survey

205

Appendix: QOUL Survey

206

Appendix: QOUL Survey

207

Appendix: QOUL Survey

208

Appendix: QOUL Survey

209

Appendix: QOUL Survey

210

Appendix: QOUL Survey

211

Appendix: QOUL Survey

212

Appendix: QOUL Survey

213

Authors Index

214

Authors Index

Bernardo, F. 12 A

Abdallah, S. 116 Birnholtz, J. P. 126 Adams, A. 124 Bissell, D. 99 Addo, I. A. 42 Biswas-Diener, R. 40 Adriaanse, C. C. M. 45 Bjornskov, C. 161 AhnAllen, J. M. 45 Blanchflower, D. G. 146 Aiello, A. 44 Bloch-Jorgensen, Z. T. 14 Albert, D. P. 64 Bocarejo, S. J. P. 97 Alfonzo, M. A. 69 Booth, A. L. 122 Allen, T. D. 127 Bowling, N. A. 124 Alshmemri, M. 118 Boyce, C. J. 146 Amérigo, M. 64 Brace, N. 149 Amole, D. 44 Brannon, D. 124 Amponsah-Tawiah, K., 127 Brief, A. P. 10 Andrews, F.M. 145 Buckley, P. 69 Aragonés, J. I. 41, 42 Burton, E. 39 Argyle, M. 10, 145 Buys, L. 46, 57 Aries, M. B., 124 Aulia, D. N. 46 C Austin, D. M. 64, 148 Avila-Palencia, I. 95 Campbell, A. 144, 145, 147 Aydin, N. 92, 96, 98, 148 Cao, X. J. 65, 69, 91, 92, 93, 94, 113, 148 Carlquist, E. 12 B Carvajal, M. J. 119, 120 Cats, O. 92, 93, 98, 99, 110 Baum, S. 44, 46 Cervero, R. 94, 100 Baumeister, R. F. 13 Chang, P. J. 161

215

Authors Index

Chapman, D. W. 64, 159 Dolan, P. 11, 146, 147, 161 Chaskin, R. J. 63 Dole, C. 124 Chaudhuri, K. 119 Dolnicar, S. 12, 148 Chen, C. F. 94 Duran, M. A. 116 Chen, L. 46 Dusansky, R. 41 Chen, L. H. 147, 161 Chevalier, A. 146 E Clark, A. E. 65, 119, 123, 126, 136, Easterlin, R. A. 147 146 Easthope, H. 41 Clark, W. A. 12, 94 Ellis, N. 121 Cohen, J. 149 Elsinga, M. 46, 59 Costa, G. 12 Erdogan, B. 124 Costa-Font, J. 39 Ettema, D. F. 61, 91, 92, 93, 94, 98, 99, 113, Cozens, P. 65, 67, 68 148 Cramm, J. M. 68 Ewing, R. 100 Crossman, A. 119 Cummins, R. A. 12, 145 Currie, G. 162 F

Fagan, C. 122 D Fellesson, M. 92, 93, 98 Félonneau, M. L. 65 Danielsson, C. 126 Fernández-Carro, C. 39, 148 Darma, P. S. 122 Fernández-Portero, C. 39, 42, 148 Dassopoulos, A. 65, 68 Ferrer-i-Carbonell, A. 146 Dawal, S. Z. M. 125 Flavin, M. 41 de Oña, J. 104 Fonner, K. L. 127 De Vos, J. 92, 93, 94, 95, 110, 113, 149 Fornara, F. 124 Dekker, K. 42, 44, 67 Foster, K. A. 63 Delhey, J. 144 Foye, C. 44 Delle Fave, A. 9 Frey, B. S. 97, 161 Di Masso, A. 66 Friman, M. 92, 93, 94, 95, 98, 104, 148 Diaz-Serrano, L. 45, 48, 71, 99, 149 Fryers, T. 115 Diener, E. 7, 8, 9, 40, 116, 146, 147 Furnham, A. 118 Ding, C. 94 Djukic, M. 125 216

Authors Index G Hulin, C. L. 115, 148 Hur, M. 66, 69, 76, 87 Galster, G. 42, 66, 68 Gao, Y. 92, 94, 96, 104, 112, 113, 127, 148, I 169 Gardner, P. G. 63 Ibem, E. O. 44, 46 Gatersleben, B. 92, 95 Iseki, H. 98 Gazioglu, S. 119 Ismail, A. 46 Gerdtham, U. G. 116 Gerhart, B. 136 J Gibson, J. J. 65 Jacobs, J. 67 Glatzer, W. 9 Jain, R. 124, 136 Goetze, R. 69 Jensen, W. A. 70 Jiang, W. 42 H Jiang, Z. 124

Haar, J. M. 117 Jiboye, A. D. 41, 46, 57 Hackman, R. 118 Joewono, T. B. 98 Hadavi, S. 65 Judge, T. A. 122, 123, 136, 141 Hagedorn, L. S. 119 Jussila, I. 45, 59 Haggan, M. 124 Hahn, E. A. 147 K Hair Jr, J. F., 52, 152 Kahneman, D. 9, 126, 159 Hall, M. 42 Karatuna, I. 121 Handy, S. 127 Kauhanen, M. 121 Harris, K. J. 124 Keul, A. G. 18 Hauret, L. 119 Kline, R. B. 82, 108 Head, A. 126 Koohsari, M. J. 63 Higgins, C. D., 93, 94, 97, 100 Kweon, B. S. 17, 66 Hincks, S. 126 Hipp, J. R. 46, 62, 63, 67 L Hooper, D. 134 Hoppock, R. 117 Lance, C. E. 10 Hosseini, D. 120 Lane, R. E. 10 Huang, Z. 44, 47 Lee, J. S. 115, 148

217

Authors Index

Lee, S. M. 66 Munshi, T. 94 Lee, S. Y. 126 Muthén, L. K. 143 Lee, T. 63 Legrain, A. 99 N Levy, P. E. 123 Nanjundeswaraswamy, T. S. 120 Li, S. 47 Newman, O. 67 Li, Z. 42 Nguyen, P. D. 126 Locke, E. A. 118, 122 Novaco, R. W. 99, 127 Loewe, N. 12 Nyberg, A. J. 136 Lottrup, L. 124 Lovejoy, K. 66, 68 Low, C. T. 16, 32, 36, 48, 51, 56, 58, 59 O

Lu, M. 44, 45, 66 Oakil, A. T. M. 96 Lynch, K. 67, 94 Ogu, V. I. 41, 43 Oh, J. 25, 44 M Oktay, D. 18 Okulicz-Kozaryn, A. 9, 16 Maat, K. 95 Olsen, J. R. 96 Macke, J. 16 Olsson, L. E. 92, 93, 126, 127, 142 Makinde, O. O. 41, 42 Orstad, S. L. 69 Marans, R. W. 13, 17, 62, 64, 144, 161 Orth, U. 146 Marcouyeux, A. 125 Ory, D. T. 112 Maslow, A. H. 41, 45, 66, 67, 118 Oud, J. H. 28, 77 Mas-Machuca, M. 117 Mattisson, K. 99 Meegan, R. 64 P

Mitchell, A. 42, 64 Pacione, M. 66 Mohit, M. A. 42, 44, 45, 47 Parkes, A. 25, 159 Mokhtarian, P. L. 92, 112 Peiser, R. B. 16 Molin, E. J. E. 128 Pekkonen, M. 47 Montero, R. 121, 122 Pietersma, S. 12 Morris, E. A. 93, 96, 97 Pinto, S. 8 Moss, S. 55, 105, 133, 134, 154 Pocock, B. 124 Müller, T. 68 Proshansky, H. M. 11 Mumford, L. 63 218

Authors Index Q Şimsek, Ö. F. 11 Sirgy, M. J. 8, 9, 11, 12, 19, 65, 69, 92, 115, Qureshi, M. I. 124 117, 118, 144, 145, 147, 148, 161, 162 Spears, S. 98 R Spencer, E. S. 119 Stanley, J. 162 Raymond, C. M. 65, 66, 68 Steg, L. 99 Redman, L. 98 Steger, M. F. 13 Rezvani, M. R. 17 Steijn, B. 119, 121 Rice, R. W. 118 Stimson, R. 13, 18 Rioux, L. 70, 125 St-Louis, E. 92, 93, 96, 97, 100, 110 Roberts-Hughes, R. 46 Stradling, S. G. 93, 98, 99 Rollings, K. A. 65 Stutzer, A. 97, 161 Rosenberg, M. J. 65 Sukriket, P. 117 Russo, M. 122 Susilo, Y. O. 92, 93, 98, 110 Ryan, R. M. 144 T S Talen, E. 64, 65, 70 Saeed, I. 117 Tao, L. 42 Safi, M. 146 Tay, L. 14 Sahin, H. 123 Teck-Hong, T. 161 Salleh, N. A. 44 Temelová, J. 42 Santos, L. D. 65 Theodori, G. L. 45 Saraji, G. N. 120 Turcotte, M. 95 Schalock, R. L. 145 Türkoğlu, H. 17 Schepers, P. 70 Türksever, A. N. E. 18 Schimmack, U. 91 Schwanen, T. 96 U Sealy–Jefferson, S. 47 Sedaghatnia, S. 17 Ullman, J. B. 55, 79, 133 Selvarajan, T. T. 119 Urry, J. 95 Sen, A. 10 Senasu, K. 115, 148 Senlier, N. 18

219

Authors Index V Washington, S. P. 55, 79, 105, 134, 154 Wasko, M. M. 124 Van Aerden, K. 123 Węziak-Białowolska, D. 62 Van Assche, J. 67 Wittmer, J. L. 121 Van Cauwenberg, J. 88 Woo, E. 11 Van den Berg, P. 31 Wyatt, T. A. 121 Van der Meer, P. H. 146 Van Praag, B. M. 145 Y Varghese, S. 117 Veenhoven, R. 10 Yavuz, N. 112 Veitch, J. A. 124 Ye, R. 93, 94, 100, 112, 113 Vera-Toscano, E. 41, 44, 47 Vestling, M. 12 Z Vischer, J. C. 125, 126 Zacher, H. 119 Visser, K. 88 Zavei, S. J. A. P. 41 Von Wirth, T. 17 Zebardast, E. 39, 148 Zhang, C. 66, 97 W Zhang, F. 39, 148

Wan, D. 104 Zhang, Z. 62, 65, 69, 148 Wang, D. 43, 45, 47, 147 Zhu, Y. 117 Wang, W. C. 43, 45, 47, 147

220

Subject Index

A Commute 92, 95, 96, 110, 112, 113, 126,

Accessibility 18, 19, 65, 69, 94, 100, 103, 127, 138, 142 104, 105, 107, 108, 109, 110, 112, 113, Conceptual Model 4, 40, 48, 52, 53, 54, 57, 126, 132, 168, 169 70, 71, 77, 79, 80, 87, 104, 105, 114, Activity 4, 8, 11, 14, 15, 19, 25, 37, 44, 61, 116, 132, 152, 153, 166, 167 63, 68, 69, 70, 71, 74, 77, 80, 81, 82, 83, Conceptualization 10, 11, 40, 52, 70, 77, 84, 85, 87, 88, 89, 95, 97, 107, 112, 113, 88, 104, 132, 144, 152 115, 116, 123, 148, 161, 162, 166, 167 Amos 53, 55, 79, 105, 107, 132, 134, 153, D 154 Design 2, 4, 21, 22, 24, 28, 34, 36, 38, 46, Architecture 25, 28, 42, 69, 73, 77, 84 47, 48, 50, 51, 52, 56, 57, 58, 59, 65, 67, 89, 140, 166, 170, 173 B Domain-Specific 2, 3, 5, 14, 91, 92, 145,

Biking 25, 37, 69, 71, 74, 76, 77, 80, 82, 83, 168 84, 85, 88, 95, 103, 108, 167 Dutch 22, 28, 32, 64, 67, 89, 114, 138, 151, Bottom-Up Approach 142, 144, 153, 168 154, 162 Built Environment 11, 14, 19, 20, 21, 26, Dwelling 17, 18, 19, 24, 25, 30, 41, 43, 45, 28, 40, 44, 65, 66, 92, 93, 94, 100, 104, 46, 47, 53, 57 113, 114, 143, 168, 172 E C Economic 7, 13, 17, 18, 28, 41, 42, 43, 44,

Cities 2, 7, 13, 18, 63, 69, 98, 165, 172 59, 60, 67, 68, 70, 116, 124, 140, 145, Cleanness 16, 25, 37, 73, 77, 84, 88, 98 163, 171

221

Eindhoven 22, 28, 29, 31, 60, 172 Histogram 48, 49, 71, 72, 100, 128, 129, Employee 117, 118, 125, 130, 138, 139, 149, 150 171, 172 Household Income 24, 28, 29, 33, 43, 54, Environment 3, 10, 11, 12, 13, 14, 17, 18, 57, 59, 60, 67, 93, 99, 135, 141, 159, 19, 20, 21, 26, 28, 40, 41, 42, 44, 65, 67, 166, 169 77, 88, 92, 93, 94, 100, 103, 104, 113, Housing Well-Being 3, 4, 14, 19, 21, 24, 114, 117, 118, 120, 121, 122, 124, 125, 143, 145, 148, 161, 166, 169, 170 130, 132, 135, 136, 138, 140, 141, 142, 143, 163, 166, 168, 170, 172 I Evaluation 9, 11, 13, 16, 17, 18, 26, 45, 47, Infrastructure 88, 89, 103, 108, 113 51, 53, 57, 64, 65, 66, 67, 68, 79, 88, 89, 96, 119, 123, 125 J

F Job Satisfaction 5, 116, 117, 118, 119, 121, 122, 123, 124, 126, 127, 131, 134, 141, Family 3, 9, 12, 14, 16, 17, 18, 19, 22, 26, 145, 168, 171, 173 27, 31, 34, 45, 100, 103, 105, 109, 110, Job Well-Being 3, 19, 21, 27, 143, 159, 162, 112, 113, 119, 121, 122, 124, 144, 145, 166, 169, 171 147, 148, 151, 152, 154, 155, 156, 157, 158, 159, 161, 162, 168, 169, 172, 173 Family Loss 159, 161, 162, 169 L

Framework 3, 4, 5, 8, 14, 15, 20, 38, 53, 62, Leisure 3, 12, 13, 14, 16, 17, 18, 19, 22, 27, 76, 77, 92, 144, 172 34, 91, 127, 144, 145, 147, 148, 151, 152, 154, 155, 158, 161, 172 G Life Domains 2, 3, 4, 5, 9, 10, 11, 12, 14, 15, 19, 21, 22, 27, 34, 122, 144, 145, 146, Gap 40, 42, 116 147, 148, 149, 152, 154, 155, 158, 159, H 161, 162, 166, 168, 170, 172 Livability9, 10, 25, 37, 67, 71, 73, 76, 77, 84 Happiness 1, 8, 9, 10, 11, 16, 17, 116, 117, Living Environment 12, 41, 163 119, 123, 142, 146, 147 Health Well-Being 14, 27, 144, 145, 147, M 159, 161, 162, 169 Mobility 13, 41, 42, 63, 94, 126, 161

222

N Preference 59, 122 Privacy 27, 125, 128, 130, 132, 135, 138, Neighborhood Satisfaction 4, 17, 18, 19, 139, 140, 142, 168, 171 25, 60, 62, 64, 65, 66, 67, 68, 69, 70, 71, Private Transport 96, 98, 99, 102, 103, 104, 72, 73, 76, 77, 79, 80, 81, 82, 83, 87, 88, 105, 107, 108, 110, 167 89, 95, 100, 104, 107, 108, 112, 113, 167 Psychology 1, 10, 11, 14, 42, 62, 65, 117, Netherlands 28, 69, 89, 110, 114, 138, 172 124, 132 Public Transport 17, 18, 26, 27, 89, 94, 95, O 96, 97, 98, 99, 102, 103, 104, 105, 107, 108, 110, 112, 130, 132, 138, 139, 140, Objective 2, 3, 5, 7, 9, 10, 13, 16, 17, 21, 167 22, 47, 50, 51, 52, 53, 56, 66, 143, 147, 162, 166, 170 Q Ownership 24, 25, 32, 34, 35, 43, 45, 47, 50, 51, 53, 56, 57, 58, 59, 60, 93, 94, Quality Of Life 1, 8, 10, 16, 17, 39, 40, 62, 112, 113, 166, 170 92, 116, 117, 148, 162, 165, 170

P R

Parking 25, 26, 31, 34, 36, 46, 47, 48, 50, Regression 4, 5, 48, 49, 50, 51, 53, 70, 71, 51, 95, 96, 102, 103, 104, 108, 109, 110, 72, 73, 76, 99, 100, 101, 102, 105, 128, 113, 167 129, 130, 132, 149, 151, 152, 153, 166 Path Analysis 4, 5, 52, 55, 128, 131, 132, Residential Satisfaction 12, 41, 42, 44, 46, 134, 142, 152, 162, 166, 168, 169 60 Path Model 53, 55, 132, 133, 154, 166 Residents 2, 8, 13, 17, 19, 21, 22, 24, 28, Peak Hours 26, 33, 37, 103, 104, 107, 110, 29, 38, 40, 42, 43, 45, 46, 47, 48, 50, 52, 112, 114, 167, 171 56, 59, 60, 61, 62, 63, 64, 66, 67, 68, 69, Perception 12, 16, 18, 40, 41, 63, 79, 94, 70, 71, 77, 87, 88, 89, 94, 95, 100, 105, 97, 123, 127, 135, 141, 142, 168, 170 107, 138, 148, 149, 152, 154, 162, 163, Personality 7, 92, 96, 112, 113, 118, 136, 165, 166, 167, 170, 171 145, 147, 148, 154, 162 Rush Hours 93, 96, 114, 167 Physical Activity 19, 25, 37, 44, 69, 70, 71, 74, 77, 80, 87, 88, 89, 107, 112, 113, 167 S Planning 7, 11, 64, 65, 89, 95, 165, 170 Policy 1, 2, 42, 163, 165, 170, 171 Safety Satisfaction 88, 97, 108

223

Salary Satisfaction 132, 135, 136, 141, 168 Travel Satisfaction 5, 26, 91, 92, 93, 94, 95, Social Satisfaction 68, 76, 77, 80, 87 96, 97, 98, 99, 100, 101, 102, 104, 107, Social Well-Being 144, 145, 148, 159, 161, 108, 110, 112, 113, 114, 127, 148, 167 169 Travel Time 26, 27, 71, 83, 93, 95, 97, 110, Society 8, 40, 97, 115, 147 112, 113, 114, 127, 132, 135, 136, 138, Socio-Demographic Variables 48, 50, 52, 142, 167, 168 57, 67, 93, 99, 104, 105, 107, 112, 113, Trip Attributes 19, 26, 93, 96, 97, 104, 110, 128, 144, 145, 149, 152, 153, 154, 155, 113, 167 158, 166, 168 Socio-Economic 28, 42, 43, 68 U Spatial 12, 28, 41, 44, 61, 63, 67, 69, 89 Unemployment 28, 145, 146, 158, 159, Stress 16, 93, 95, 96, 97, 99, 120, 123, 124, 161, 162, 169, 171 125, 127, 146 Urban Life Domains 3, 4, 15, 19, 21, 166, Subjective Well-Being 2, 3, 4, 5, 8, 14, 21, 168, 172 28, 42, 92, 117, 144, 145, 146, 147, 162, Urbanism 16, 65, 95 165, 172 Survey 3, 8, 18, 21, 22, 23, 24, 27, 33, 35, 38, 77, 136, 199 W

Walking 25, 31, 37, 68, 70, 74, 77, 79, 80, T 82, 83, 85, 88, 95, 98, 103, 108, 167 Work Location 19, 27, 117, 124, 127, 128, Theory 3, 9, 10, 11, 14, 16, 19, 41, 42, 45, 132, 134, 135, 136, 168 65, 89, 97, 118, 136, 144, 153, 168 Work Satisfaction 19, 27, 38, 115, 116, 117, Time Schedule 26, 70, 89 118, 119, 124, 127, 128, 129, 130, 131, Traffic Congestion 17, 25, 26, 62, 79, 95, 132, 134, 135, 136, 138, 139, 140, 141, 96, 97, 110, 171 142, 168, 172 Transport Satisfaction 95, 102, 104, 105, Workload 27, 119, 120, 121, 122, 123, 107, 108, 110, 130, 138, 139 127, 130, 132, 135, 136, 140, 141, 168 Travel Cost 26, 93, 96, 97 Workspace 19, 27, 38, 124, 125, 128, 130, Travel Mode 33, 93, 94, 95, 98, 104, 112, 132, 134, 135, 136, 138, 139, 140, 141, 113, 114, 167 168

224

Summary

Modelling and Measuring Quality of Urban Life: Housing, Neighborhood, Transport and Job

Research on Quality of Life (QOL) has a long history in social science. It tends to focus on specific domains and how people’s satisfaction about that domain is related to their quality of life. Considering that most of the world’s population is living in cities and that the share of the urban population continues to increase leading to various problematic negative externalities, it is not surprising that the concept of QOL re-emerged on the academic agenda, also of engineering disciplines as the Quality of Urban Life (QOUL) concept. The concept quickly found its way into main research streams and rapidly gained momentum. Particularly in urban planning, researchers attempted to investigate the effect of various aspects of cities on citizens’ well-being, and achieved considerable progress in developing methodological and theoretical approaches concerning QOUL. However, for a long time, these studies were based on aggregate data, focusing on aspects such as crime and economic challenges without considering the role of subjective judgments of residents. Borrowing subjective evaluation methods from social and behavioral sciences, urban planning researchers expanded the scope and instrumentality of this line of research. Now, it is evident that in order to understand QOUL, it is essential to investigate how residents judge urban places. However, despite these promising methodological advances in QOUL studies, there are still very few studies that provide a comprehensive framework to measure residents’ wellbeing in urban settings. The majority of studies focuses on a single urban life domain. Although these studies have been instrumental in optimizing the determinants of urban life, they may not reflect the extent by which various life domains influence each other and overall QOUL. In fact, by ignoring other life domains, the estimates of quality of life and the marginal effects of the attributes of the considered life domain may be seriously biased, potentially leading to wrong policy and design recommendations.

225

This limitation of most prior urban planning and transportation research motivated the present PhD study. The main aim is to develop a comprehensive model of QOUL, expanding and integrating concepts that have been suggested over the last few decades to predict subjective well-being of residents in urban settings. Thus, the main objective of this dissertation is to estimate the relationships between QOUL and satisfaction in different life domains with an explicit focus on the role of the built environment, while controlling for other life domains.

In order to achieve these aims, an extensive literature review of the study of QOL in different life domains was conducted. The insights from the review provided the theoretical foundation for the development of the conceptual framework of this thesis. As subjective well-being is a multi-dimensional concept involving many life domains, a domain-specific approach based on bottom-up spillover theory was used to develop the initial model. Moreover, overall QOUL is hypothesized to be a function of the person- place-activity relationships, and thereby the main life domains were derived from the concept. Regarding place, housing well-being and neighborhood well-being were included in the model as both house and neighborhood are key living places. For the movement between places, transport was included, while work, social, shopping and leisure activities were added to the model to represent daily activities. Finally, health, family and income were included in the model.

To operationalize the conceptual framework, data were collected through an online questionnaire distributed among households in four neighborhoods in the city of Eindhoven in the summer and autumn of 2016. The questionnaire included questions about subjective well-being in the four selected domains of housing, neighborhood, transport and job, plus questions about financial situation, health, leisure, social life, family life and socio-demographic characteristics. Considering the influence of personal attributes and the characteristics of the residential area on QOUL evaluations, the study areas were chosen such as to reflect differences in socio-economic position and differences in the characteristics of the built environment.

Different types of models are used to analyze the complex relationships embedded in the developed theoretical framework. For each domain, considering the nature of the dependent variable, interrelationships among various indicators are estimated using regression analysis, path models and structural equation models. Results indicate the main determinants of QOUL in each life domain. The estimation results of the models lead to the following conclusions. First, in the housing well-being

226

model, satisfaction with ownership, type of house and design are the most important predictors of overall housing satisfaction. Regarding the objective characteristics of the house, lack of safety facilities, and living in low cost rental houses significantly and negatively influence residential satisfaction of households. Among the socio- demographic variables, length of residency and household income influence residential satisfaction positively, mediated through ownership.

Second, in the neighborhood well-being model, social aspects such as neighborhood attachment and safety turn out to be more important than environmental and physical aspects. From neighborhood services, only grocery shopping is found to be influential. Both walking and biking conditions are found to significantly influence overall neighborhood satisfaction, with satisfaction with walking being more important. Among the socio-demographic variables, being single influences neighborhood satisfaction negatively, mediated through aspects such as neighborhood attachment.

Third, in transport well-being model, traveling in peak hours is found to be the main predictor of travel satisfaction. More specifically, rush hours avoidance has a positive and significant effect on travel satisfaction. Moreover, mode choice directly influences commuting satisfaction. Accessibility satisfaction is a strong predictor of transport well-being. Considering that socio-demographic characteristics of individuals directly influence commuting choices, which in turn influence travel satisfaction, results show that younger people and males are more likely to travel during peak hours, and therefore they have lower travel satisfaction compared with other groups. As a group of employed people in our sample avoids peak hour commuting, which leads to higher travel satisfaction, we can see the positive role of teleworking and work flexibility on travel satisfaction.

Fourth, in the job well-being model, the results indicate a strong influence of work conditions satisfaction on overall work satisfaction. Satisfaction with quality of work from the work condition category is found to have a strong influence on work satisfaction. Satisfaction with privacy in the work space and satisfaction with travel time to work are other important factors influencing job satisfaction.

Finally, the results of estimating overall QOUL reveal interesting findings as well. Income well-being, health well-being and social well-being are the most important predictors of QOUL, respectively. This means that physical domains have less impact compared to the aforementioned domains, which is in line with the findings of psychological and happiness research. Moreover, transport well-being is the only

227

domain with an indirect influence on QOUL. Considering the multiple associations among the domains, there are many more findings related to the interrelationships among the domains, such as the impact of transport well-being on neighborhood well- being, or the effect of neighborhood well-being on house well-being. All in all, as the four major urban domains that we investigated have significant influence on QOUL, we can conclude that our results confirm the relevance of the conceptual framework underlying this thesis.

228

Curriculum Vitae

After completing her Master degree in Architecture in 2008 at Faculty of Architecture of Mashhad University, Lida worked in architecture consultancies and firms as a landscape designer and architect. Also, she taught some courses related to theory and practice of architecture and urban design at different institutes. Moreover, she collaborated with Iranian architecture magazines in writing and editing articles and was active in organizing talks and seminars. Later, she moved to Norway and worked as a designer and developer for three years. Having interest in interdisciplinary research and gaining knowledge in urban studies, she started conducting research on urban regeneration as a guest researcher at Delft University of Technology. In October 2013, she joined the Urban Planning Group at TU/E to pursue her PhD research and started working on the topic “Quality Of Urban Life”. During the PhD study, she broadened her knowledge of many urban related topics such as housing, neighborhood, transportation and job and gained vast knowledge on theory and analysis of urban systems. She has presented her work at many conferences and workshops such as AESOP, ERES and INGRID. Her research interests include urban planning, quality of life and urban design.

229

230