Transportation Research Part F: Traffic Psychology and Behaviour
Clearing the path to transcend barriers to walking: Analysis of associations between perceptions and walking behaviour
Graphical abstract
Introduction
In response to major threats such as climate change, exclusion, or sedentary lifestyle-related illnesses, cities are increasingly focused on liveability, health, or equality of access (C40 Cities., 2018, Glazener and Khreis, 2019, UN-HABITAT., 2014, UN-HABITAT, 2016, World Health Organization. (n.d.). Let’s be active, the global action plan on physical activity, 2018). Everyday walking is gaining traction globally as a policy goal (Auckland Auckland Council, 2018b, Giles-Corti, 2017, Lowe et al., 2015; UN UN DESA, 2016, UN-HABITAT., 2014, UN-HABITAT, 2016), given its contributions to equity of access (Burdett, 2018, Gibson et al., 2012, NZIER., 2014, Rose et al., 2009), participation (Bigonnesse et al., 2018, Eisenberg et al., 2017, Fomiatti et al., 2014, Hoenig et al., 2003, Mindell, 2017), physical activity (Alidoust and Bosman, 2015, Annear et al., 2014, Badland, 2007, Eisenberg et al., 2017, Haselwandter et al., 2015, Webber et al., 2019), urban economic efficiency (Davis and Golly, 2017, McCann, 2009), and lower greenhouse gas emissions (C40 Cities., 2018, United Nations, 2015). The potential of retrofit and better urban design for encouraging and enabling walking are now well understood (Gehl, 2010, Gunn et al., 2017, Macmillan et al., 2020, Speck, 2012). The question authorities around the world battle with is how to improve urban environments to make the biggest difference, bearing in mind pragmatics such as economic and time constraints (Burdett, 2018, The Landscape Architecture Foundation, 2016, UN-HABITAT, 2016).
Walkability research has made important progress in understanding walking behaviour and barriers to walking. It is now widely accepted that walking is simultaneously influenced by the urban environment (UE) – encompassing buildings, greenery, and traffic, but also by individual, organisational or community factors (socio-ecological framework) (Alfonzo, 2005, Forsyth, 2015, Sallis et al., 2016, Sallis, 2009). There is, however, no consensus on the relative importance of diverse UE characteristics on walking (Alfonzo, 2005, Badland, 2007, Forsyth, 2015, Franckx, 2017, Kerr et al., 2016, McCormack and Shiell, 2011). Promising theoretical developments have been made and some testing has been done (Alfonzo, 2005, Buckley et al., 2016, Mehta, 2008). While it is unclear if a unique and robust model for understanding walking behaviour exists, it is important to also understand what is not walkable and might shift people to use another mode or avoid trips within walking distance (Alfonzo, 2005, Buckley et al., 2016, Forsyth, 2015).
A recent systematic umbrella review examined the development of the conceptual framework linking the built environment and walking behaviours (Bozovic, Hinckson, & Smith, 2020). Briefly, the Social Model of Walkability posits that the relationship between the UE and an individual’s walking behaviour is moderated conjointly by (1) people’s perceptions of their environment (namely their perceptions of the satisfaction of their walking needs: how feasible, accessible, safe, comfortable or pleasant a trip is); (2) individual characteristics (e.g. disability1, constraints, preferences or available alternatives); (3) trip purpose, and (4) internal motivations. The review concluded that little attention has been given to environmental perceptions in both research and in the modernist approaches to urban design and transport planning.
This study builds on the findings of the realised umbrella review (Bozovic et al., 2020), exploring the associations between perceptions, individual characteristics and walking behaviour. This exploration considers Tamaki Makaurau-Auckland, Aotearoa-New Zealand, a city of 1.66 million residents (2018) (Auckland Council, 2018a). Auckland’s transport infrastructure and low density are comparable to those of other car-oriented cities (Nunns, 2014).
Auckland Transport is the agency charged with urban transport planning and operations. Data from Auckland Transport's Active Modes online survey (AT survey) were used. The AT survey aims to understand behaviours, attitudes and perceptions of different modes of travel, over time (TRA, 2017), capturing perceptions of the walking environments, as well as a vast array of possible motivators and deterrents, and travel behaviour.
The aim of this study is to compare the importance of (1) perceptions of the walking environments and namely those perceptions relative to the satisfaction of the walking needs; (2) public transport use; and (3) individual characteristics in the prediction of walking levels. The individual variables used as inputs all relate conceptually to walking, however the novelty of the present approach is to consider them simultaneously against the walking behaviour. The assumption is that all three aspects play a role in predicting walking levels, the focus being on their relative importance.
Section snippets
Setting and data
Auckland’s development has had a strong focus on traffic infrastructure and urban sprawl (Auckland Council, 2018b; Gehl Architects, 2010). Jan Gehl saw a fantastic location and natural environments but described the city as “a rush hour 'traffic machine'”, referring to a car-centric design (Gehl Architects, 2010). The car-centric design contributes to the high rates of pedestrian deaths and serious injuries (Howard, 2018), social isolation (Rose et al., 2009), reduced affordability for the end
Pairwise associations
Multiple pairwise associations were noted between perceptions, motivations, individual characteristics, and walking behaviour. Each of the 41 variables were significantly associated with 12–33 other variables. Walking levels and safety at night as a barrier were both associated with 33 other variables. The chi-squared test results are presented in the Table 3 below.
The identified multicollinearity confirmed the strategy of using machine learning for modelling walking as an outcome based on
Discussion
The study assessed the relative importance of users’ perceptions, motivations, and individual characteristics in relation to walking levels. Walking levels were predominantly explained by perceived qualities of walking within the transport system. Surprisingly, the importance of the perception of living too far for walking to be practical was marginal for predicting the walking levels (3.6% for PT users, 1% for non-users and 0.6% for the total population). We observed a multiplicity of
Conclusion
The findings provide four main take-aways for both research and the practice: (1) users’ perceptions of their environments need to be better understood and linked to objective aspects of the walking environment; (2) walking needs to be considered within the transport system – as a complement to public transport or an alternative to other modes; (3) it is crucial to embrace the diversity of users, examining how different constraints (e.g. having difficulties walking, seeing or hearing) might
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgments
Tamara Bozovic is supported by the Auckland University of Technology Doctoral Scholarship, and Melody Smith is supported by a Sir Charles Hercus Health Research Council of New Zealand Research Fellowship (grant number 17/013).
The authors would like to acknowledge Auckland Transport for having made the anonymised survey data available for this research.
References (116)
‘Complacent car addicts’ or ‘aspiring environmentalists’? Identifying travel behaviour segments using attitude theory
Transport Policy
(2005)- et al.
Why do people walk? Role of the built environment and state of development of a social model of walkability
Travel Behaviour and Society
(2020) - et al.
The business case for walking—Counting walking to make walking count in Auckland, New Zealand (breakout presentation)
Journal of Transport & Health
(2017) Built environmental correlates of physical activity in China: A review
Preventive Medicine Reports
(2016)- et al.
Does the built environment moderate the relationship between having a disability and lower levels of physical activity? A systematic review
Preventive Medicine
(2017) - et al.
Disability and dignity-enabling home environments
Social Science & Medicine
(2012) - et al.
A multi-level approach to travel mode choice – How person characteristics and situation specific aspects determine car use in a student sample
Transportation Research Part F: Traffic Psychology and Behaviour
(2011) - et al.
Housing and transport expenditure: Socio-spatial indicators of affordability in Auckland
Cities
(2014) - et al.
Use of science to guide city planning policy and practice: How to achieve healthy and sustainable future cities
The Lancet
(2016) To walk or not to walk? The hierarchy of walking needs
Environment and Behavior
(2005)