Citation: Kamelifar, M.J.; Ranjbarnia,
B.; Masoumi, H. The Determinants of
Walking Behavior before and during
COVID-19 in Middle-East and North
Africa: Evidence from Tabriz, Iran.
Sustainability 2022,14, 3923. https://
doi.org/10.3390/su14073923
Academic Editor: JoséCarmelo
Adsuar Sala
Received: 31 January 2022
Accepted: 15 March 2022
Published: 26 March 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
sustainability
Article
The Determinants of Walking Behavior before and during
COVID-19 in Middle-East and North Africa: Evidence from
Tabriz, Iran
Mohammad Javad Kamelifar 1, Behzad Ranjbarnia 2and Houshmand Masoumi 3,4,*
1Tabriz Metropolis Municipality, Tabriz 5174643119, Iran; [email protected]
2Department of Geography and Urban Planning, Faculty of Planning and Environmental Sciences,
3Center for Technology and Society, Technische Universität Berlin, Kaiserin-Augusta-Allee 104,
10553 Berlin, Germany
4Department of Transport and Supply Chain Management, College of Business and Economics,
University of Johannesburg, Johannesburg 2006, South Africa
*Correspondence: [email protected]
Abstract:
To support the global strategy to raise public health through walking among adults,
we added the evidence on predictors of walking behavior in the Middle East and North Africa
(MENA) region by emphasizing the mediator—COVID-19. During the COVID-19 outbreak, public
restrictions to encompass the spread of the disease have disrupted normal daily lifestyles, including
physical activity and sedentary behavior. It was proposed that tremendous changes have occurred on
predictors of physical activity in general and walking behavior in particular for three types of walking,
including commute, non-commute, and social walking compared to pre-COVID-19 time. This study
aimed to identify the determinants of the walking types mentioned above, including subjective and
objective variables before COVID-19, and compare them during the COVID-19 period in a sample
from Iran, which has not yet been addressed in previous research. Adults (N = 603) finalized an
online survey between June 5 and July 15, 2021. This group reported their individual/socioeconomic
locations (e.g., home/work) and perception features before and during COVID-19. The paper
developed six Binary Logistic (BL) regression models, with two models for each walking type
(commute, non-commute, and social walking). For commute trips before COVID-19, the findings
showed that factors including BMI, residential duration, p. (perceived) neighborhood type, p.
distance to public transport stations and job/university places, p. sidewalks quality, p. facilities
attractiveness, p. existence of shortcut routes, commute distance, building density and distance to
public transport were correlated with commute walking. At the same time, such associations were not
observed for BMI, p. distance to public transport and job/university places, p. facilities attractiveness,
building density, and distance to public transport during COVID-19. The variables include age,
possession of a driving license, number of family members, p. neighborhood type, p. distance to
grocery, restaurant, parking, and mall, p. existence of sidewalks, land-use mix, and distance to
public transport indicated correlations with non-commute before COVID-19. However, p. distance to
groceries and malls and the p. existence of sidewalks did not correlate with non-commute walking
during COVID-19. Ultimately for social walking, age and income variables, and the considerable
proportions of subjective variables (e.g., p. distance to services/land-uses, security, etc.), health status
and building density were correlated with social walking before COVID-19. Nevertheless, most of
the mentioned variables did not explicitly correlate with social walking during COVID-19. As for the
implication of our study, apparently, special actions will be needed by urban authorities to encourage
adults to enhance their walkability levels by fully considering both objective and subjective indicators
and walking types, which will result in healthier lifestyles.
Keywords:
transportation planning; urban travel; walking behavior; COVID-19; Middle-East and
North Africa
Sustainability 2022,14, 3923. https://doi.org/10.3390/su14073923 https://www.mdpi.com/journal/sustainability
Sustainability 2022,14, 3923 2 of 20
1. Introduction
The Coronavirus Disease 2019 (COVID-19) pandemic has infected millions world-
wide [
1
,
2
] and impacted many more. Because of its transmission from person to person [
3
],
many developed and developing countries (e.g., USA, England, Turkey, Iran) implemented
interventions to induce mobility restrictions and forced their citizens to stay at home (i.e.,
confinement), to reduce the transmission rate and prevent health services from being over-
whelmed, unless shopping for basic necessities, attending to medical issues, engaging in
one form of exercise a day (e.g., a run, walk, or cycle), or if employed in an essential job [
4
].
These facts changed the lifestyle of most people, engaging them in less physical activity
and more sedentary behavior than pre-COVID-19 times due to imposed restrictions [
5
].
For example, according to a report from Google Mobility (2020), as of 23 June, substantial
decreases occurred in access to services such as retail and recreation (
−
48%), supermarket
and pharmacy (
−
10%), public transport (
−
48%), and workplaces (
−
51%) during the pan-
demic in the United Kingdom. Considering that physical inactivity is one of the key risk
indicators for chronic illness and all-cause death [
6
], such limitations on human mobility
may have negative health effects in the long run. Consequently, research on the effects of
the pandemic on changed patterns of physical activity predictors can be considered a global
public health need [
7
]. Such need is also considered more substantiated in developing
countries than developed ones due to insufficient health data and infrastructures.
Accordingly, examining the factors that facilitate walking during quarantine and
lockdown periods and comparing it with pre-COVID-19 time may assist in developing
strategic initiatives aimed at offsetting negative consequences [
8
,
9
]. Based on the literature,
various factors influence physical activity, including objective [
10
,
11
] and subjective [
12
,
13
]
measures. The emerging literature in recent years has tried to investigate the relationship
between objective or subjective neighborhood characteristics and active transportation
behavior [
14
–
18
]. These studies support realizing the determinants of walking behavior
better and an inclusive analytical framework for walkability research. However, the
volume of these studies in scope and geography is rare in developing countries, and
more investigations need to be conducted. In addition, employing an integrated approach
(using objective, and subjective factors simultaneously) is significantly scarce in previous
studies. To the best of our knowledge, no study has yet compared and modeled the
determinants of walking behavior before and during the pandemic; there are research gaps
in the mentioned areas.
There is a need to grasp better the role of independent variables in shaping various
types of walking in different contexts in terms of the mediator, COVID-19 outbreak. This
article has taken three modes of walking, including commute, non-commute, and social
walking (with a partner/friend), as dependent variables. This approach will help planners
and stakeholders have a comprehensive and accurate understanding of the objective
and subjective determinants of walking in metropolitan areas of developing countries.
Accordingly, the research objectives are to undertake micro-scale studies to: 1—identify
the determinants of walking behavior of adults for job-traveling, non-commute, and social
walking purposes in the large cities of developing countries; 2—compare the walking
behavior determinants for commute, non-commute, and social walking purposes before
and during the outbreak of COVID-19.
The rest of the article contains the following units. Section 2reviews the existing
literature on the concept of walking behavior determinants of various types, the impact of
the COVID-19 pandemic on walking behavior, and these studies’ situation in developing
countries. In Section 3, we briefly discuss the employed methods in this paper. Subse-
quently, in Section 4, we present a summary of results from statistical methods. Section 5
discusses the model similarities and differences of walking behavior determinants before
and during the outbreak of COVID-19 and compares the findings with those conducted for
developed countries. Ultimately, Section 6includes the conclusion.
Sustainability 2022,14, 3923 3 of 20
2. Literature Review
2.1. Objective and Subjective Measures of Walkability
Walkability is the supporting characteristic of complete, sustainable, and healthy cities.
Walkable surroundings can encourage physical activity through determined or entertaining
walking, subsequently adding social value to the public and acquiring healthy life [19,20].
Accordingly, walking behavior is influenced by ample factors addressing, environmental
(objective) and psychological (subjective) contexts. Opportunity is conceptualized as
encompassing both the subjective (perceptive wellbeing, security, neighborhood type, etc.)
and objective (e.g., environment, distance, variety) contexts that constrain or enable a
behavior. Additionally, individual characteristics should not be waived for these analyses.
Some studies have considered relationships between walking and primarily socioeco-
nomic factors, including age, income, gender, education, BMI, work status, marital status,
having a dog, having school-aged children, social distancing practices, physical/mental
health condition, infection status, and previous activity level [21–26].
Objective indicators of walkability have been predominantly employed in walkability
studies more as commute walking determinants. Of importance, certain environment
attributes are related to increased walking by community members. Previous studies
have incorporated explicit measures of environmental attributes (e.g., [
27
–
29
]), or have
developed composite factors for exploring the relation to walking behavior [
30
,
31
]. Among
environmental factors, land-use and connectivity measures are two main categories whose
effects have been studied previously [
32
,
33
]. These variables calculate the built neighbor-
hood’s design, density, and diversity quantitatively [
34
,
35
]. The most common indicators
in these studies list factors including building and population density [
36
], land-use mix
diversity [
11
,
37
,
38
], land-use mix access [
37
], street connectivity [
30
,
38
], visual dimensions
of a place such as lighting [
39
], etc. These indices have been proven useful in explaining
walkability in different metropolitan areas [40,41].
Despite the importance and rich background of objective measures in explaining
walkability, others have emphasized the subjective variables in interpreting behavioral con-
sequences of environmental walkability [
30
]. Yet, a big body of the study has admitted that
perceived walkability is equal to the built environment variables [
42
,
43
] if not more signifi-
cant when considering predictors of the walking behavior [
44
]. The subjective dimension
favors individual values and beliefs shaped by social contexts and personal experiences [
45
],
meaning perspectives and experiences of environments are unique between individuals,
particularly those living in different spaces. Many studies have demonstrated the merit
influence of subjective factors, including safety, neighborhood type, wellbeing, etc., on
walking behavior [
46
–
48
]. The study results presented by French et al. (2013) indicated that
perceived behavioral control was the key determinant of walking, mostly in non-commute
walking [
49
]. Lee (2016) explains that neighborhood perception and safety are important
factors determining walkability [
50
,
51
]. Factors such as lifestyle and wellbeing, beliefs, and
personality could influence travel behavior [
52
,
53
]. In some cases such as in non-commute
walking, “as mentioned”, their effect has been reported greater than the impact of the built
environment [
54
]. However, a review of the effects of workplace relocation on commuting
mode change showed that perceptive variables were poor predictors of commuting mode
change when relocating workplaces [55].
Apart from debates on both commute and non-commute walking, another type is re-
garded as social walking or walking with a companion. This type of walking has been inves-
tigated very shortly compared to those mentioned. In a universal context, very few studies
have investigated the determinants of companion walking. In terms of predictors of walk-
ing accompanied by someone, a large body of studies has focused on health
status [56,57]
,
socio-economics (e.g., gender, income, age) [
58
,
59
], individual
perceptions [60,61]
, and
environmental factors (very limited factors including accessibility) [
62
,
63
]. The integrated
approach is very short and inconsistent, which needs to be more thoroughly investigated.
Overall, the literature review regarding both objective, and subjective measures shows
that although walking behavior has been defined with different travel variables, most
Sustainability 2022,14, 3923 4 of 20
studies have only used one measure for their analysis and do not follow an integrated ap-
proach. The review also shows that environmental factors are the most widely investigated
characteristics among various explanatory factors, and the effects of subjective factors on
walking behavior are limited.
2.2. Impact of COVID-19 Pandemic on Walking Behavior
Undoubtedly the imposed mobility restrictions that were executed to decrease the
spread of the COVID-19 pandemic have affected walking behavior. Still, these variations’
extent and spatial-psychological characteristics have not been thoroughly confirmed yet.
Much less is known about the differential impacts of COVID-19 response measures on
the walking behavior of populations. Although there is an evolving body of literature
measuring the possible impact of COVID-19 upon walking and physical activity, Sallis et al.,
emphasized the requirement for more investigations on this topic [
7
]. In 2020, researchers
from different nations (e.g., Belgium, Canada, Greece, USA, Australia) measured walking
behavior changes resulting from their diverse COVID-19 restrictions. Summarizing these
research studies, physical activity was lower during COVID-19 restrictions than before
restrictions were put in place [
64
,
65
]. People’s fear of being in public places is also expected
to reduce outdoor activity [
23
,
66
]. However, other research suggests that the pattern of
behavior may be more complex. Cheval et al. (2021) reported a rise of about 10 min in
walking and moderate physical activity and an increase of roughly 75 min in sedentary
behavior between adults in Switzerland and France. While some activities (e.g., walking
for commuting) have been reduced, other activities (e.g., working out indoors) have gone
up [
67
]. However, this assessment similarly endorses the substantial between-individual
differences in the impact of the restrictions on physical activity. Research has yet to
compare potential changes in physical activity before and during COVID-19 and how
walking behavior determinants may have changed.
Changes in social life and the daily routine (i.e., loneliness) and the imposed stress on
individuals (insecurity about their health or financial consequences) have also been reported
to affect sleep patterns negatively [
68
,
69
]. Despite the attempts to indicate the changes
and determinants in physical activity [
67
,
70
], the evidence is still limited. In addition, very
limited studies and comparisons have been accomplished on determinants of walking
behavior before and during the lockdown. Further, all previous studies on this topic have
used subjective tools; therefore, objective and quantitative ways to monitor population
behavior are required to determine the integrated impact of the various objective and
subjective indicators on walking behavior to suggest strategies for possible confinements
soon. In sum, this integrative approach is particularly relevant in the context of the
COVID-19 crisis, which has caused sudden changes in people’s work, family, and living
environment. This necessitates performing comparative studies to understand changes in
determinants of walking behavior.
To investigate the relationships between both environmental (objective), perceptive
(subjective) factors as an independent variable, and walking (commute and non-commute
walking and having a companion) as a dependent variable before and during the COVID-19,
we adopted the same model comparison approach.
2.3. The Condition in Developing Countries
Investigation on walking behavior determinants in a universal context is sparse and
inconsistent. Previous studies regarding objective and subjective factors on walking were
mainly conducted in developed countries such as the United States [
71
], Canada [
72
], the
United Kingdom [
73
], and Australia [
74
]. Studies examining the perceptions and attitudes
that impact walking behavior for neighborhood travel are scarce in developing countries.
These nations have different social, historical, cultural, and environmental characteristics
relative to developed countries, and thus the psychological factors influencing people’s
walking behavior in these countries may also differ from those in developed countries [
75
].
Sustainability 2022,14, 3923 5 of 20
New evidence from developing countries can help us refine our understanding of the
attitudinal and perceptual factors that determine people’s level of walking.
The previous works in developing countries are even more dramatically scarce. For ex-
ample, most of these studies have been conducted in Turkey [
76
], Brazil [
77
], and
Iran [78,79]
.
For example, some studies in Iran have been conducted on children/adolescents’ walking
behavior, particularly in some cities, including Tehran, Yazd, Rasht, and Kerman [
80
,
81
]
with no consideration for the integrated approach. These studies analyzed a wide range
of policy-sensitive variables. They found that walking time to school, car ownership, and
safety concerns negatively impacted active traveling to school. In the area of adults/seniors,
these studies are even rare. In a more recent study by Hatamzadeh and Hosseinzadeh (2020),
they concluded that the top priority policy that may lead to a higher probability of choosing
walking as a mode of transportation among adults is to plan for higher mixed-use develop-
ments that could afford more accessibility and make the neighborhood more stimulating for
the adults which could itself raise the inclination to walk and as a result increase the various
group’s engagement [82].
From what was mentioned in the literature review, these gaps can be identified:
Apart from existing contextual gaps in developing countries (North Africa and Middle-
East in particular) regarding walking behavior determinants, the lack of employing the
integrated approach (using both objective and subjective determinants of walking behavior
simultaneously) in most universal studies is observed. Further, to the best of our knowl-
edge, the impact of the COVID-19 pandemic as a mediator on walking behavior, along
with comparing it with the before-COVID-19 period, has not yet been addressed in both
developed and developing countries. Ultimately, taking into consideration the different
modes of walking (commute, non-commute walking, and social walking) as dependent
variables is innovative in these countries, which is what this research aims to address.
This paper aims to answer the following questions to identify both subjective and
objective determinants of walking behavior in the understudied context of MENA:
(1) What factors determine the walking behavior of adults for the commute, non-commute,
and social walking purposes during COVID-19 in the large cities of the MENA region?
(2)
Have the determinants of walking behavior for commute, non-commute, and social
purposes changed after the outbreak of COVID-19 in the MENA region?
We hypothesize the existence of a huge difference between walking behavior deter-
minants before and during the COVID-19 pandemic. In addition, we hypothesize that
subjective factors have the most significant impact on changing the walking behavior
after the outbreak of COVID-19 in the large cities of the MENA region. Social walking is
affected by the pandemic more than other types of non-commute and commute trips. This
paper assumes that both subjective and objective determinants of walking behavior for
commute, non-commute, and social purposes have different patterns before and during the
outbreak of COVID-19 in the MENA region. The effects of subjective measures on walking
behavior have changed more radically after the pandemic compared to objective factors.
The subjective factors including perceived accessibility to neighborhood facilities, subjec-
tive walkability, perceived attractiveness of facilities, perceived security, and subjective
wellbeing have had stronger significant differences between pre-COVID-19 times and after
the pandemic. The objective measures that have had significant differences before and after
the pandemic include accessibility to public transport stations and construction density.
3. Materials and Methods
3.1. Case Study
Tabriz is the fourth largest city of Iran and one of its metropolitan towns with approx-
imately 1.6 million citizens (2021). It is located in the northwest of Iran and is the most
populous and largest city. This city is regarded as one of the top industrialized cities within
the country of Iran with a high rate of pollution as a result of cars and factories emissions.
This city is regarded as one of the top industrialized cities within Iran, with a high rate
of pollution due to cars and factories emissions. Like any other big town in Iran, this city
Sustainability 2022,14, 3923 6 of 20
has witnessed rapid transformation during the last 25 years. The city was selected as the
first healthy city of Iran in 2020 [
83
]. The geographical characteristics of Tabriz are listed in
Table 1.
Table 1. Geographical Characteristics of Tabriz.
Key Features Tabriz
City (km2)325 km2(125 sq. mi)
Urban Area (km2)512 km2(198 sq. mi)
Divisions 10 Districts
Urban Planning Governance Tabriz Municipality
Urban Transportation System Governance
Organization of transportation and Traffic
Availability of city-wide Urban Development Plan Yes (2013)
Availability of city-wide Strategic Transportation Master Plan Yes (2019)
Changes in urbanization after the emergence of modernism in urban planning have
destroyed the traditional elements and structures of the city. Moreover, functional zoning
has replaced the organic system and functional diversity of city and urban neighborhoods.
Consequently, walk-friendly neighborhoods altered, and the overall walkability of the city
has decreased dramatically. It is to say that the establishment of the BRT system started
about 15 years ago to lessen the heavy traffic of its main street, and recently one of the
metro lines has started serving people. However, this city suffers from a lack of sufficient
and high-qualified public transportation systems moving many citizens to use private cars
in their commutes.
3.2. Data and Variables
This study is based on data collected from an online questionnaire survey conducted
between 5 June to mid-July 2021 and secondary data using a 1.2000 GIS map (objective).
Respondents provided detailed information about their individual characteristics, percep-
tions of their neighborhood, mode choice for commute, non-commute, social trips, and
health status. Of the 1090 people who started the survey, only 668 respondents completed
enough questions to be used in these analyses. Some were removed during the modeling
procedure due to item non-response (603 remained). The total sample size was 603 and is
representative of Tabriz city, according to Cochran (1963). This paper comprises exploratory
research aimed to identify walking behavior determinants and provide insights into their
variation before and during the outbreak of COVID-19 in large Iranian cities, and as an
example, for large MENA cities.
Socioeconomic characteristics, mode choice, participants’ perception, health status,
and built environment (objective) variables are presented in Table 2. The variables are
classified into binary, continuous, and categorical variables. Continuous variables are
age, monthly income, BMI (body mass index), residential duration, number of family
members, commute distance, land-use mix, distance to public transport, and building
density. Categorical variables include neighborhood type, access to different services,
neighborhood environment, facilities attractiveness, walkable places, security, wellbeing,
and health status.
Sustainability 2022,14, 3923 7 of 20
Table 2. Socioeconomic, subjective, and objective variables.
Variables Description and Coding
Age Continuous
Gender Female = 1, Male = 0
Education
Diploma and Undergraduate = 1, Bachelor = 2, Master = 3, PhD
and higher = 4
Income Continuous
Possession of Driving License Yes = 1, No = 0
BMI Continuous
Number of family members Continuous
Residential Duration Continuous
Neighborhood Type Less than 2 floors = 1, 2 to 6 floors = 2, 6 to 10 floors = 3,
10 to 20 floors = 4, more than 20 floors = 5
Cul-de-sac From 1 = very little to 5 = very much
Distance to Different Services Less than 5 min = 1, 5 to 10 min = 2, 10 to 20 min =3,
20 to 30 min = 4, More than 30 min = 5
Perceptive Neighborhood Environment (1. Existence of trees
across the street, 2. Existence of architecturally attractive
buildings and houses, 3. Existence of attractive scenery for
walking, 4. Suitable slope of streets for walking, 5. Existence of
Suitable urban furniture and benches at short distances)
From 1 = very little to 5 = very much
Perceptive Walkable Places (1. Existence of sidewalks, 2.
Separation of street from sidewalk by green spaces, 3. Existence
of shortcut routs)
From very little = 1 to very much = 5
Overall Quality of Sidewalks (Width, attractiveness, quality of
materials and ups and downs) From very bad = 1 to very good = 5
Facilities Attractiveness
No existence = 1, Not attractive at all = 2, Not very attractive = 3,
Medium = 4, Acceptable Attractiveness = 5, Very attractive = 6
Perceptive Security (1. The streets of the neighborhood are not
well lit at night, 2. Due to the crime rate, our neighborhood is
not secure enough, 3. There is a lot of traffic on the streets
around our neighborhood making walking difficult and
unpleasant, 4. There are no pedestrian crossing signs on the
busy streets of our neighborhood, 5. The streets do not have
speed bumps)
From Strongly disagree = 1 to Strongly agree = 4
Wellbeing and Health Status (1. Feeling of Depression, 2.
Feeling of Anxiety, 3. Feeling of Being Energetic, 4. Feeling
cheerful and cool, 5. Emotionally stable and confident, 6.
Overall Health status)
From Very Low = 1 to Very High = 10
Objective Built Environment (Commute distance- Distance to
public transport- Street length, Land use mix, Number of
Intersections, Building & Population density)
Continuous
Travel Mode Choice Walking = 1, Other modes (e.g., Private car, Public transport,
Metro and . . . ) = 0
3.3. Analysis Methods
We developed six binary logistic (BL) regression models using travel mode choices
(walking and other types of trips) as the dependent variables to determine the influencing
factors. Three BL regression models were generated for commute, non-commute, and social
walking before COVID-19; we then repeated this process for the mentioned walking types
during the outbreak of COVID-19. The Six BL models for walking behavior determinants
Sustainability 2022,14, 3923 8 of 20
showed how different they were before and during the pandemic. Through the six models,
the determinants of walking behavior were established based on the perceptions of residents
(subjective), and the built environment (objective). The first round of BL models used
42 variables as independent variables. Variables were then eliminated from the BL models
based on the highest p-value. This process was repeated until a suitable model was achieved
based on significant variables and a higher value of Nagelkerke’s R
2
. An Omnibus test
reveals the validity of the BL models with significant variables (p-values of less than 0.05)
and higher Nagelkerke’s R2values.
4. Results
4.1. Descriptive Statistics
The survey respondents were 603 residents of Tabriz city. In connection with gender,
59.2% of respondents were women, while 40.8% were men. Although they came from
diverse age groups, the least represented group was the 18s, and the majority of respondents
were aged between 28 and 40 at the time of the survey. The detailed statistics regarding
gender and age can be found in Figure 1.
Figure 1. Frequency of Age and Gender in the sample.
Regarding the quality of sidewalks, about 18% of respondents considered the quality
to be higher than moderate. Moreover, only 6.3% of participants found the quality of
facilities in their neighborhoods very attractive, while 21.8% supposed that the facilities in
their neighborhoods were not absorbing at all. In terms of mode choices for commute and
non-commute trips, it should be noted that around 13.5% and 22.5% of contributors choose
walking as their modes of transport, respectively. Table 3shows the frequency of different
mode choices for commute and non-commute trips before and after the COVID-19 outbreak.
Sustainability 2022,14, 3923 9 of 20
Table 3. Commute and non-commute trips.
Commute Trips Non-Commute Trips
Before COVID-19
Pandemic
During COVID-19
Pandemic Before COVID-19
Pandemic
During COVID-19
Pandemic
Category Frequency Percent Frequency Percent Frequency Percent Frequency Percent
Walking 81 13.4% 108 17.9% 134 22.2% 135 22.4%
Private Car 248 41.1% 280 46.6% 299 49.6% 348 57.7
Bus 85 14.1% 35 5.8% 60 10% 23 3.8%
Taxi 68 11.3% 81 13.4% 53 8.8% 38 6.3%
Taxi Apps 36 6% 38 6.3% 28 4.6% 35 5.8%
Metro 11 1.8% 7 1.2% 9 1.5% 5 0.8%
Organizational Service
42 7% 42 7% 5 0.8% 8 1.3%
Bicycle 13 2.2% 12 2% 13 2.2% 11 1.8%
Motorbike 1 0.2% 0 0 0 0 0 0
Missing Data 18 3% 0 0 2 0.3% 0 0
Total 603 100% 603 100% 603 100% 603 100%
4.2. Model Fit
We generated six BL models for this research. Here, we present the final models after
the elimination of insignificant variables. The implications for further research are then
discussed subsequently.
4.2.1. The Impact of the Built Environment and Individual Perception on Commute
Walking Behavior before and during COVID-19
The best model for walking behavior determinants for commute trips before COVID-
19 was generated with 11 significant variables after running 22 models. Subsequently, these
variables were employed similarly for commute trips during COVID-19 to be comparable.
The first model’s dependent variable is the dummy variable of pre-COVID-19 travel mode
choices. We divided the respondent’s answers into two categories of walking and other
modes of travel. Having the highest p-value, the following variables were each omitted
from the model: age, gender, income, education, job, possessing of a driver’s license,
number of family members, disability status, distance to different services (e.g., distance
to parks, malls, bank, etc.), perceptive neighborhood environment (1. the existence of
trees across the street, 2. the existence of architecturally attractive buildings and houses,
3. the existence of attractive scenery for walking, 4. suitable slope of streets for walking,
5. the existence of Suitable urban furniture and benches at short distances), perceptive
walkable places (1. the existence of sidewalks, 2. separation of the street from the sidewalk
by green spaces), perceptive security (1. the streets of the neighborhood are not well lit at
night, 2. due to the crime rate, our neighborhood is not secure enough, 3. there is a lot of
traffic on the streets around our neighborhood, making walking difficult and unpleasant,
4. There are no pedestrian crossing signs on the busy streets of our neighborhood, 5. The
streets do not have speed bumps), Perceptive Wellbeing and Health Status (1. feeling of
depression, 2. feeling of anxiety, 3. feeling of being energetic, 4. feeling cheerful and cool,
5. emotionally stable and confident, 6. overall health status) and objective built environment
(1. street length, 2. population density, 3. distance to public transport, 4. land use mix,
5. Intersection). The perceived distance to one’s job/university has a significant negative
correlation with walking behavior, while the real distance (commute distance) has less
significance. It is the same for the perceived distance to public transport stations and real
distance to public transport. The probable explanation for this is that perceptions are more
influential in commute and long-distance walking than objective factors. There is a positive
and highly significant correlation between shortcut routes and walking for commute due
to time travel reduction, which is an effective factor for commute trips.
Furthermore, perceived neighborhood type is highly significant in the model due to
its substantial role in walking behavior across the entire population. As can be understood
from Table 3, residents with higher rates of BMI tend to choose walking as their mode
of transport for commuting, probably due to its impact on losing weight and staying fit.
Furthermore, facility attractiveness and overall quality of sidewalks positively correlate
Sustainability 2022,14, 3923 10 of 20
with walking for commute trips. Besides this, despite insignificancy, residential duration
and building density were kept in the model to improve the results. Nagelkerke’s R
2
is
0.28, correctly covering 90.7% of the variables.
As mentioned, we generated our first model to recognize walking behavior predictors
for commute trips before the outbreak of COVID-19. Subsequently, we developed our
second model to compare the determinants of walking behavior before and during the
COVID-19 spread. With the aim of comparison, the independent variables are exactly the
identified variables in the first model, while the dependent variable is the dummy variable
of during-COVID19 travel mode choices. As Table 4indicates, most variables lose their
significance after the pandemic, which can be interpreted to mean that residents are more
likely to choose other modes of transport during COVID-19 or prefer not to commute due
to remote working. Among the variables, perceptive distance to job/university, commute
distance, neighborhood types, and shortcut routes are still significant. This likely happens
due to their crucial impact on utilitarian walking.
Table 4. Binary logistic models for commute walking behavior pre and after COVID-19.
Pre-COVID-19 During COVID-19
Variable/Measure Wald B Beta pWald B Beta p
BMI Before/During COVID-19 5.390 0.132 1.141 0.020 0.995 0.042 1.043 0.319
Residential Duration 2.654 −0.026 0.974 0.103 2.977 −0.024 0.976 0.084
P. (perceived) Neighborhood Type 7.845 −0.385 0.680 0.005 5.476 −0.262 0.770 0.019
P. Distance to Public Transportation Stations 5.062 −0.366 0.694 0.024 1.991 −0.194 0.824 0.158
Distance to Job/University 9.493 −0.548 0.330 <0.001 10.311 −0.495 0.641 <0.001
Sidewalks overall quality 3.601 0.371 1.449 0.058 0.690 0.139 1.149 0.406
Facilities Attractiveness 5.294 0.403 1.496 0.021 0.045 −0.029 0.971 0.832
Walkable Places 3 (existence of shortcut routs)
9.341 0.529 1.697 0.002 11.664 0.516 1.675 0.001
Commute Distance 5.253 −0.493 0.611 0.022 4.034 −0.363 0.696 0.045
Building Density 0.874 0.158 1.171 0.350 0.935 0.138 1.148 0.333
Distance to P. Transport 3.729 2.092 8.098 0.053 1.066 0.771 2.162 0.302
Constant 2.930 −3.716 0.024 0.087 0.090 −0.510 0.600 0.764
Omnibus Test of Model Coefficients Chi-Square pChi-Square p
55.352 <0.0001 41.677 0.000
Hosmer and Lemeshow Test Chi-Square pChi-Square p
3.703 0.883 10.922 0.206
Model Summary −2 Log likelihood
Nagelkerke
R2
Percentage
correct −2 Log likelihood
Nagelkerke
R2
Percentage
correct
195.642 0.281 90.7 258.288 0.190 87.9
4.2.2. The Impact of Built Environment and Individual Perception on Non-Commute
Walking Behavior before and during COVID-19
In this step, after running 25 models, the greatest model for pre-COVID-19 walking
behavior determinants for non-commute trips was developed using the following highly
significant (p-value of less than 0.05) and marginally significant (0.05 < p-value < 0.1)
variables: age, possession of a driving license, number of family members, neighborhood
type, distance to grocery, distance to restaurant, distance to parking, distance to mall, the
existence of sidewalks, land-use mix, and distance to public transport. Age and possession
of a driving license had a negative and highly significant correlation with pre-COVID-
19 walking behavior. The same results are seen for the distance to mall, distance to
restaurant, and distance to public transport (objective) variables, which means that before
the pandemic, with the increasing perceived distance to malls/restaurants and increasing
objective distance to public transport, the rate of walking is reduced dramatically among
the residents. On the other hand, some factors, including the number of family members,
neighborhood type, distance to parking, and land-use mix variables, had a highly positive
association with pre-COVID-19 walking behavior. The highly positive association between
land use mix and walking behavior can be interpreted to mean that as land use mix
increases, the likelihood that a resident chooses to walk for non-commute trips also rises.
In this regard, sidewalks and distance to grocery variables are marginally significant and
positively connected to pre-COVID-19 walking behavior. Then, non-commute walking
behavior determinants were compared before COVID-19 and during COVID-19 pandemic
periods based on the previous step procedure. As Table 5indicates, except for three
Sustainability 2022,14, 3923 11 of 20
variables, namely the existence of sidewalks, grocery distance, and restaurant distance,
other variables are still significant. The distance to restaurants variable loses its significance
probably due to lockdown regulations during the pandemic in which authorities enforced
restaurant closure. In addition, no determination rate was observed for distance to grocery
stores during COVID-19, which can probably be interpreted that residents could not
commute for daily shopping. The Omnibus and Hosmer–Lemeshow tests prove that the
model is significant; Nagelkerke’s R
2
and the correct percentage of variables in the model
point out that the model is an acceptably good fit (Table 4).
Table 5. Binary logistic models for non-commute walking behavior pre and after COVID-19.
Pre-COVID-19 During COVID-19
Variable/Measure Wald B Beta pWald B Beta p
Age 4.578 −0.022 0.979 0.032 5.715 −0.024 0.976 0.017
Possession of Driving License 12.161 −0.941 0.390 <0.001 12.512 −0.943 0.389 <0.001
Number of family members 5.581 0.204 1.226 0.018 5.320 0.199 1.220 0.021
Neighborhood Type 4.805 0.337 1.401 0.028 7.681 0.432 1.540 0.006
Distance to Grocery 3.073 −0.289 0.736 0.080 1.839 −0.220 0.746 0.175
Distance to Restaurant 4.095 −0.281 0.755 0.043 1.872 −0.187 0.829 0.171
Distance to Parking 6.931 0.245 1.277 0.008 3.309 0.168 1.182 0.069
Distance to Mall 9.863 −0.893 0.409 0.002 8.994 −0.845 0.430 0.003
Walkable Places 1 (Existence of sidewalks)
2.741 0.165 1.180 0.098 0.483 0.068 1.071 0.487
Land-use mix 9.368 0.128 1.137 0.002 2.958 0.070 1.073 0.085
Distance to Public Transport 6.779 −0.994 0.432 0.009 5.595 −0.984 0.398 0.018
Constant 4.585 1.931 6.894 0.032 7.054 2.381 10.817 0.008
Omnibus Test of Model Coefficients Chi-Square pChi-Square p
50.845 <0.0001 40.705 <0.0001
Hosmer and Lemeshow Test Chi-Square pChi-Square p
8.615 0.376 8.324 0.402
Model Summary −2 Log likelihood
Nagelkerke
R2
Percentage
correct −2 Log likelihood
Nagelkerke
R2Percentage
correct
536.293 0.134 78.9 546.432 0.108 78.4
4.2.3. The Impact of Built Environment and Individual Perception on Companionship in
Walking before and during COVID-19
The binary logistic model for companionship in walking pre-COVID-19 was generated
after removing the insignificant variables with higher p-values. The data in Table 5show
that perceived distance to grocery stores, restaurants, parks, and malls, along with facilities
attractiveness and overall health status, have a positive and highly significant association
with companionship in walking. On the other hand, although there are significant as-
sociations between sidewalks, shortcut routes, security 1, security 2, a feeling of anxiety,
depression, and building density with social walking, these correlations are negative. The
negative association between security 1,2 (streets are not well lit at night, the neighborhood
is not secure), the existence of sidewalks variables and companionship in walking can
probably be due to the security matters, which means having a companion in walking can
give a sense of security and confidence to pedestrians. There are four marginally significant
variables: income, cul-de-sac, distance to job/university, the existence of shortcut routes,
and being energetic, meaning they have a slight impact on companionship in walking.
Compared to companionship in walking before the pandemic, the companionship
after the pandemic has decreased dramatically, probably due to the infectious nature of
COVID-19 and the resulting outcomes. The dominance of subjective variables in this type
of walking is promising compared to the two other walking modes. Table 6represents the
details of our model in terms of social walking.
Sustainability 2022,14, 3923 12 of 20
Table 6. Binary logistic models for Companionship in walking pre and after COVID-19.
Pre-COVID-19 During COVID-19
Variable/Measure Wald B Beta pWald B Beta p
Income 3.057 −0.145 0.865 0.08 2.886 −0.129 0.879 0.089
Cul-de-sac 3.032 0.278 1.321 0.082 0.557 −0.106 0.899 0.455
Distance to Grocery 5.336 0.493 1.611 0.021 3.812 −0.368 0.692 0.051
Distance to Restaurant 6.180 0.313 1.368 0.013 3.297 0.204 1.227 0.069
Distance to Park 4.799 0.329 1.72 0.028 3.563 −0.249 0.78 0.059
Distance to Mall 10.894 1.351 3.86 0.001 7.056 0.978 2.658 0.008
Distance to Job/Uni 3.407 0.197 1.821 0.065 2.641 −0.163 0.85 0.104
Sidewalks overall quality 4.920 0.254 0.776 0.027 0.096 −0.032 0.968 0.756
Facilities Attractiveness 9.355 0.388 1.475 0.002 4.441 0.239 1.27 0.035
Walkable Places 1 Existence of sidewalks 4.790 −0.247 0.781 0.029 1.309 −0.123 0.885 0.253
Walkable Places 3 (existence of shortcut routs)
3.569 −0.381 0.683 0.059 0.846 −0.167 0.846 0.358
Security 1 (streets are not well lit at night) 7.234 −0.406 0.667 0.007 1.899 −0.18 0.835 0.168
Security 2 (neighborhood is not secure) 5.608 −0.404 0.667 0.018 0.038 −0.028 0.972 0.846
Feeling of Depression 10.020 −0.16 0.852 0.002 0.061 −0.011 0.989 0.804
Being energetic 3.219 0.116 1.123 0.073 2.644 0.112 1.118 0.104
Feeling of Anxiety 10.888 −0.200 0.819 0.001 1.281 −0.07 0.933 0.258
Overall Health Status 5.128 0.307 1.359 0.024 8.094 0.396 1.486 0.004
Building Density 4.881 −0.267 0.765 0.027 0.975 −0.096 0.908 0.323
Constant 2.012 1.818 6.161 0.156 0.001 0.034 1.034 0.978
Omnibus Test of Model Coefficients Chi-Square pChi-Square p
82.973 <0.0001 36.056 0.007
Hosmer and Lemeshow Test Chi-Square pChi-Square p
4.576 0.802 8.453 0.391
Model Summary −2 Log likelihood
Nagelkerke
R2
Percentage
correct −2 Log likelihood
Nagelkerke
R2Percentage
correct
354.670 0.304 71.8 400.541 0.143 66.3
5. Discussion
This research examines and compares the correlations between objective neighborhood
variables, perceived built environment, and walking of three different types, commute,
non-commute, and social walking before and during the COVID-19 pandemic in Tabriz.
We found that most respondents had reduced their levels of overall walking during the
lockdown resulting from fear of infection and disease anxiety. A large body of literature
mostly emphasizes non-commute walking. There is less evidence of investigation on the
correlation of various indicators with commute and social walking. To the best of our
knowledge, this is the first analysis to investigate the changes in adults’ walking behavior
before and during the COVID-19 pandemic considering the differences in various types of
walking. Our findings driven from binary logistic regression demonstrate the important
variances in the impact of different variables on the modes of walking behavior according
to the perceived built environment and physical neighborhood factors, which are in line
with some pre-conducted studies [84,85].
In our first and second model (commute determinants before and during COVID-19),
socioeconomic factors indicated the least correlation when predicting walking before and
during the pandemic, which contradicts some previous studies [
86
,
87
]. The role of these
factors is promising. The variable BMI showed positive correlations with commute before
COVID-19, which is consistent with some previous findings [
85
,
88
–
90
]. Those with higher
BMI were likely inclined to walk to reach their jobs for health reasons. However, that
significance was not observed during COVID-19. It is probable that COVID-19 anxiety and
infection fear prevented them from walking. Regarding subjective variables, significant
negative correlations were seen between perceived neighborhood type and distance to
different services/land uses (e.g., Public Transport and Job/University) with commute
walking before COVID-19. It means that by increasing the perceived compactness of the
neighborhood as a result of perceived high building density and enhancing distance to job
places, the tendency for commute declined before COVID-19, which confirms that residents
did prefer to commute using other transport modes, mostly private cars in perceived
dense and distanced areas. This would be consistent with some conducted studies [
91
–
93
].
Nevertheless, such a correlation did not exist during COVID-19 for the factor of “distance
to public transport”. This is probably because, during COVID-19, individuals did not select
public transport as their commute mode choice. Further, remote work can be regarded
Sustainability 2022,14, 3923 13 of 20
as another reason. Other subjective factors such as sidewalks’ overall quality, facilities
attractiveness, and shortcut routes, indicated positive correlations with commute before
COVID-19. While, except for the variable of “the existence of shortcuts”, the other two
variables were not significant determinants during COVID-19.
Our third and fourth models for non-commute walking determinants before and
during COVID-19 included more socioeconomic determinant variables signifying the im-
portance of individual characteristics’ role in determining the use of non-commute walking
rather than commute. The age variable showed no considerable difference compared to
the pre-COVID-19 period. The significance is even higher during COVID-19. Young adults
are likely more motivated to do neighborhood walking during COVID-19, which is in line
with previous studies [
21
,
94
] in China and Jordan. In terms of having a driving license,
despite enhancing private car usage in COVID-19 times, no considerable difference has
taken place after the COVID-19 outbreak meaning those who possess a driving license have
a lesser tendency to engage in non-commute walking. Accordingly, some studies have
drawn similar results in the USA, Jamaica, etc. [
95
]. Similarly, the same positive correlations
were obtained for some family members. Similar to our first model for commute walking,
the variable of neighborhood type (perceived compact area) showed a correlation with
non-commute walking. However, this correlation is negative, which probably indicates
that perceived compact areas have high attractiveness for non-commute walking, enabling
residents to meet their daily needs before and during COVID-19. In this regard, there
are some consistent investigations [
10
,
11
,
37
,
38
]. Apart from it, negative associations were
observed between perceived distance to non-commute targets (e.g., Grocery, Restaurant,
and mall) with walking before COVID-19. In contrast, no associations existed between
perceived distance to groceries and restaurants during COVID-19. People have probably
been using other modes rather than walking to provide their needs from grocery shops. In
the case of restaurants, due to the pandemic restrictions, most of them had no inner service.
Perceived distance to parking showed a positive correlation with walking in both pre and
post-COVID-19, meaning that individuals prefer walking mode rather than driving in cases
they have no closer access to the parking. Believing in the existence of sidewalks, preferably
well-designed ones, is regarded as an important determinant in non-commute walking
based on the findings of this study before COVID-19. At the same time, such correlation
was not found during COVID-19, which indicates the impact of the pandemic in pushing
people toward a sedentary lifestyle. Ultimately, two objective factors including land use
mix (in line with studies of Christian et al., 2011; Boakye-Dankwa et al., 2019) [
40
,
96
] and
distance to public transport (in line with investigations of Newmann and Kenworthy, 1989;
Holden and Norland, 2005) [
97
,
98
] explain non-commute walking with similar results
before and during COVID-19 periods. Sufficient diversity of land uses probably increases
the chance of non-commute walking. Higher distance to public transport acts as a barrier
against walking on both periods. The obtained results demonstrate the consistency of our
study’s objective and subjective measures regarding non-commute walking.
Our fifth and sixth models (social walking determinants before and during COVID-19)
explain the predictors of walking with a companion. Here, the income is negatively associ-
ated with walking, which means most higher-income groups do not tend to walk with a
partner; that is likely these groups prefer to use private cars once they have a partner due to
their dominance in Iranian cities. This output is consistent with the study in [
57
]. It showed
that individuals from sparsely-populated regions and lower-income categories were over-
represented in larger walking groups. However, the outcomes of the majority of studies
conducted in westernized countries contrast our results [
59
,
99
]. Apart from this, perceived
Cul-de-sac and distance to various services/land uses (distance to Grocery, Restaurant,
Park, Mall, Job/Uni) positively correlate with social walking. This can be interpreted as
indicating that the perceived security matters or lack of street lighting in the neighbor-
hoods create a sense of anxiety, making a partner/friend accompany the individuals before
COVID-19. However, during COVID-19, such associations were not seen in two factors of
Cul-de-sac and distance to Job/Uni. It is likely that by decreasing various types of walking
Sustainability 2022,14, 3923 14 of 20
during COVID-19, choosing other modes of transport has increased. Perceived quality of
sidewalks and some other factors including facilities attractiveness, being energetic, and
overall health status were positively associated with walking accompanied by a partner
before COVID-19, while during COVID-19, some of these factors such as sidewalk quality
and being energetic did not show any correlations with social walking. The interpretation
is that the overall walking in each type has decreased during the pandemic.
On the other hand, sidewalk and shortcut routes, lack of street lighting, neighborhood
security, and feeling of depression and anxiety indicated negative correlations with com-
panionship walking before COVID-19. Although, none of the mentioned variables showed
any correlations during COVID-19. Lastly, the only significant objective determinant was
building density before COVID-19, which turned out to be uncorrelated during COVID-19.
Therefore, in these models, people’s perceptions are vital in predicting social walking before
COVID-19.
The results of this study are based on six binary logistic models; they indicate that the
different characteristics in both periods before and during COVID-19 affect various types
of walking. In other words, huge differences were observed in determinants of walking
behavior in periods before and during COVID-19. Among the different types of walking,
social walking was massively affected by the COVID-19. The role of subjective variables is
vital in this type of walking. However, the significance of the correlated subjective variables
was lost during the pandemic. Therefore, the COVID-19 pandemic is regarded as a strong
mediator in predicting walking behavior. The differences between the distribution of both
subjective and objective determinants of walking behavior were also proven. In other
words, the model of each type of walking was made by bringing together the particular
factors that created different patterns. Such diversity of walking behavior determinants
needs the attention of urban authorities. In our study, the obtained results showed consis-
tency and inconsistency of objective and subjective variables for different types of walking.
In this regard, Gebel et al. (2009) and Gebel et al. (2011) state that the perceived neigh-
borhood walkability is not essentially the same as objective neighborhood walkability,
which is consistent with some outputs of this study in non-commute walking [
100
,
101
]. It
seems that some individuals are likely to misperceive their environment as having low
or high walkability. The fact that perceived behavioral control was pointed out as the
strongest predictor of walking corroborates previous research findings [
102
,
103
]. Despite
the importance of objective variables, especially in non-commute walking, subjective data
should complement one another to produce a more holistic understanding of walkability
and walking behavior. In our study, the effects of subjective factors including “perceived
accessibility to neighborhood facilities, subjective walkability, perceived attractiveness of
facilities, perceived security, and subjective wellbeing” have radically changed compared
to the objective variables’ impact after the COVID-19 outbreak. Accompanied by subjective
variables, the two objective factors of “accessibility to public transport stations and building
density” have had significant differences. These differences are due to the establishment
of new metro stations, as well as urban regeneration and expansion schemes. Lastly, in-
cluding socioeconomic circumstances and health status or motivation into understandings
of walkability investigations are vital for interpreting the interplay between individuals
and their environment [
104
]. The benefit of integrating various data, “which is regarded
as the strength of this research”, is that they may guide more cost-effective interventions.
For example, rather than encouraging the restructuring of neighborhood networks, inhab-
itants may request for strategic organizing of cross-walks or the introduction of paved
walking trails.
We found that those who care more about their health and gas emissions from cars are
more likely to walk for a commute. Seemingly these groups are more aware of the health
benefits of walking and reducing car usage. Increased awareness and knowledge regarding
active transportation’s environmental and health benefits may be important in promoting
active transportation. Besides that, rationally distanced establishment of public transport
stations is considered another strategy for those whose workplaces are rather far away. For
Sustainability 2022,14, 3923 15 of 20
non-commute walking, one important policy implication of this study concerns improving
neighborhood design, especially to encourage including more services and land uses that
support the strategy of land-use mix. Incorporating elements such as green space, shade,
benches, and recreational facilities could enhance all ages groups’ comfort and aesthetic
experience. Finally, the important intervention for social walking could be the same with
non-commute walking and creating open spaces for more social interactions that directly
impact the negative feelings of individuals and make it desirable to walk with a partner in
the neighborhood. These implications emphasize before COVID-19, as this phenomenon
will disappear due to human knowledge.
The strengths of this study, apart from using different types of determinants including
both objective and subjective factors, included the timeliness of the survey during the
COVID-19 period, the collection of retrospective data to capture the period before the
pandemic, and the use of standardized self-report walking measures. However, there were
some limitations. The data were collected during the pandemic, which forced us to do
online sampling with 603 participants. A larger sample size could run an additional reliable
set of data for modeling various types of walking. Further, the sample mostly contained
younger adults from middle-income households.
6. Conclusions
This paper aimed to identify walking behavior determinants in walking of three types,
Commute, Non-commute, and Social walking, and compare them for two different periods,
before and during the outbreak of the COVID-19 pandemic. It was conducted among
adults in one large Iranian city (Tabriz) as an example of a developing country context. In
conclusion, numerous lessons are apparent from this study.
First, the volume of walking has increased in both types of commute and non-commute
trips during COVID-19 compared to before COVID-19 (due to the decrease in the use of
public transport), which contradicts with some of the studies carried out in developed and
developing countries [
94
,
105
], for example, US and China. Similarly, in Mena countries
(Egypt, Jordan, United Arab Emirates, Kuwait, Bahrain, Saudi Arabia, Oman, Qatar, Yemen,
Syria, Palestine, Algeria, Morocco, Libya, Tunisia, Iraq, and Sudan), a study conducted by
Abouzid et al. (2021) confirmed that walking levels in these countries have declined (before
29.2%; during 20.3%) [
106
]. However, in line with the findings of this study, there are some
examples in both developed and developing nations that indicate an increase in walking
levels [
107
,
108
], for example, Bangladesh and Canada. Commute trips in Iran showed a
4.5% increase while the proportion of non-commute trips remained constant before and
during the COVID-19 pandemic (before: 22.2%; During: 22.4%). In other words, the results
are highly localized in both developing and developed countries.
Second, the determinant factors of walking behavior during COVID-19 in Tabriz
included:
−
Commute walking: perceived neighborhood type, perceived distance to job/university,
perceived sidewalks overall quality, perceived existence of shortcuts and commute
distance.
−
Non-commute walking: perceived neighborhood type, perceived distance to parking,
perceived distance to mall, distance to public transport.
−
Social walking: perceived distance to grocery, perceived distance to restaurant, per-
ceived distance to park, perceived distance to mall, perceived attractiveness of facili-
ties, and perceived overall health status. The details can be found in Tables 4–6.
Third, although generally, we found that some objective and subjective indicators
appear to be significant determinants of the walking behavior before COVID-19, the
majority of these determinants have lost their significance due to the emergence of COVID-
19. Thus, as a strong mediator, the COVID-19 plays a fundamental role in walking habits.
Accordingly, it was proven that there are huge differences between walking behavior
determinants before and during COVID-19.
Sustainability 2022,14, 3923 16 of 20
Fourth, although both objective and subjective variables appear to have merit influ-
ence on walking behavior either in commute or non-commute trips, the social walking
determinants clearly indicate the dominance of subjective variables rather than objective
measures. Therefore, it was proven that subjective variables have the most significant
impact on the walking behavior of adults in large cities of the MENA region.
Fifth, despite the fact that various types of walking have been largely affected by the
COVID-19 pandemic, this impact on social walking was even more compared to commute
and non-commute walking. Accordingly, 10 out of 16 factors lost their significance after
the pandemic. Hence, it was proved that social walking has been affected more during
COVID-19.
Sixth, the effects of subjective factors including “perceived accessibility to facilities, sub-
jective walkability, perceived attractiveness of facilities, perceived security, and subjective
wellbeing” on various types of walking have dramatically changed after the emergence of
the COVID-19 pandemic. Such alterations have also been considered in objective measures
including accessibility to public transport and building density.
Overall, the results obtained from this study comprehensively add new evidence to
the existing literature, showing that influential variables act remarkably differently before
and during the COVID-19 outbreak. In other words, the mediator, COVID-19, highlights
important differences related to the determinants of walking behavior. In addition, the
findings provide an opportunity to offer some policies and design implications to similar
cities in Mena or other developing regions to encourage adults toward active living. Thus,
there is a substantial need for more integrated and holistic research in developing and
developed countries that enables academicians to compare and help stakeholders and
policymakers accurately plan to enhance walkable communities.
Author Contributions:
Conceptualization, H.M., M.J.K. and B.R.; methodology, M.J.K. and B.R.;
software, M.J.K. and B.R.; validation, H.M.; formal analysis, M.J.K. and B.R.; investigation, H.M.,
M.J.K. and B.R.; data curation, H.M.; writing—original draft preparation, H.M., M.J.K. and B.R.;
writing—review and editing, H.M., M.J.K. and B.R.; visualization, H.M.; supervision, H.M.; project
administration, H.M.; funding acquisition, H.M. All authors have read and agreed to the published
version of the manuscript.
Funding:
This work was supported by the German Research Foundation and the Open Access
Publication Fund of Technische Universität Berlin.
Institutional Review Board Statement:
Ethical review and approval were waived for this study. No
intervention has been applied in this study, and the data used in the analyses were derived only from
surveys using a questionnaire, so an ethical review has not been conducted.
Informed Consent Statement:
Written informed consent has not been obtained from the respondents
of the interviews of this study, due to the cultural conditions, in which data were collected.
Data Availability Statement:
At the time of the publication of this paper, the data have not been
shared on a repository.
Acknowledgments:
We acknowledge support from the German Research Foundation and the Open
Access Publication Fund of Technische Universität Berlin.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
Sohrabi, C.; Alsafi, Z.; O’Neill, N.; Khan, M.; Kerwan, A.; Al-Jabir, A.; Iosifidis, C.; Agha, R. World Health Organization Declares
Global Emergency: A Review of the 2019 Novel Coronavirus (COVID-19). Int. J. Surg. 2020,76, 71–76. [CrossRef] [PubMed]
2.
Spence, J.C.; Rhodes, R.E.; McCurdy, A.; Mangan, A.; Hopkins, D.; Mummery, W.K. Determinants of Physical Activity among
Adults in the United Kingdom during the COVID-19 Pandemic: The DUK-COVID Study. Br. J. Health Psychol.
2021
,26, 588–605.
[CrossRef] [PubMed]
3.
Kampf, G.; Todt, D.; Pfaender, S.; Steinmann, E. Persistence of Coronaviruses on Inanimate Surfaces and Their Inactivation with
Biocidal Agents. J. Hosp. Infect. 2020,104, 246–251. [CrossRef]
Sustainability 2022,14, 3923 17 of 20
4.
Michie, S.; West, R.; Rogers, M.B.; Bonell, C.; Rubin, G.J.; Amlôt, R. Reducing SARS-CoV-2 Transmission in the UK: A Behavioural
Science Approach to Identifying Options for Increasing Adherence to Social Distancing and Shielding Vulnerable People. Br. J.
Health Psychol. 2020,25, 945–956. [CrossRef] [PubMed]
5.
Musselwhite, C.; Avineri, E.; Susilo, Y. Editorial JTH 16 –The Coronavirus Disease COVID-19 and Implications for Transport and
Health. J. Transp. Health 2020,16, 100853. [CrossRef]
6.
Lee, I.-M.; Shiroma, E.J.; Lobelo, F.; Puska, P.; Blair, S.N.; Katzmarzyk, P.T. Effect of Physical Inactivity on Major Non-
Communicable Diseases Worldwide: An Analysis of Burden of Disease and Life Expectancy. Lancet
2012
,380, 219–229. [CrossRef]
7.
Sallis, J.F.; Adlakha, D.; Oyeyemi, A.; Salvo, D. An International Physical Activity and Public Health Research Agenda to Inform
Coronavirus Disease-2019 Policies and Practices. J. Sport Health Sci. 2020,9, 328–334. [CrossRef]
8.
Arden, M.A.; Chilcot, J. Health Psychology and the Coronavirus (COVID-19) Global Pandemic: A Call for Research. Br. J. Health
Psychol. 2020,25, 231–232. [CrossRef]
9.
Holmes, E.A.; O’Connor, R.C.; Perry, V.H.; Tracey, I.; Wessely, S.; Arseneault, L.; Ballard, C.; Christensen, H.; Cohen Silver, R.;
Everall, I.; et al. Multidisciplinary Research Priorities for the COVID-19 Pandemic: A Call for Action for Mental Health Science.
Lancet Psychiatry 2020,7, 547–560. [CrossRef]
10.
Lee, C.; Moudon, A.V. Correlates of Walking for Transportation or Recreation Purposes. J. Phys. Act. Health
2006
,3, S77–S98.
[CrossRef]
11.
DuToit, L.; Cerin, E.; Leslie, E.; Owen, N. Does Walking in the Neighbourhood Enhance Local Sociability? Urban Stud.
2007
,44,
1677–1695. [CrossRef]
12.
Ambrey, C.; Bitzios, M. Demystifying Residents’ Walking Behaviors: Active Transport in South East Queensland, Australia. Int. J.
Sustain. Transp. 2018,12, 737–752. [CrossRef]
13.
Kwan, R.Y.; Lee, D.; Lee, P.H.; Tse, M.; Cheung, D.S.; Thiamwong, L.; Choi, K.-S. Effects of an MHealth Brisk Walking Intervention
on Increasing Physical Activity in Older People with Cognitive Frailty: Pilot Randomized Controlled Trial. JMIR MHealth UHealth
2020,8, e16596. [CrossRef] [PubMed]
14.
Liu, J.; Zhou, J.; Xiao, L. Built Environment Correlates of Walking for Transportation: Differences between Commuting and
Non-Commuting Trips. J. Transp. Land Use 2021,14, 1129–1148. [CrossRef]
15.
Singleton, P.A. Walking (and Cycling) to Well-Being: Modal and Other Determinants of Subjective Well-Being during the
Commute. Travel Behav. Soc. 2019,16, 249–261. [CrossRef]
16.
De Vos, J.; Schwanen, T.; Van Acker, V.; Witlox, F. Travel and Subjective Well-Being: A Focus on Findings, Methods and Future
Research Needs. Transp. Rev. 2013,33, 421–442. [CrossRef]
17.
De Vos, J. Analysing the Effect of Trip Satisfaction on Satisfaction with the Leisure Activity at the Destination of the Trip, in
Relationship with Life Satisfaction. Transportation 2019,46, 623–645. [CrossRef]
18.
Nordbakke, S.; Schwanen, T. Well-Being and Mobility: A Theoretical Framework and Literature Review Focusing on Older
People. Mobilities 2014,9, 104–129. [CrossRef]
19. Black, K. Health and Aging-in-Place: Implications for Community Practice. J. Community Pract. 2008,16, 79–95. [CrossRef]
20.
Leyden, K.M. Social Capital and the Built Environment: The Importance of Walkable Neighborhoods. Am. J. Public Health
2003
,
93, 1546–1551. [CrossRef]
21.
Alomari, M.A.; Khabour, O.F.; Alzoubi, K.H. Changes in Physical Activity and Sedentary Behavior Amid Confinement: The
BKSQ-COVID-19 Project. Risk Manag. Healthc. Policy 2020,13, 1757–1764. [CrossRef] [PubMed]
22.
Ingram, J.; Maciejewski, G.; Hand, C.J. Changes in Diet, Sleep, and Physical Activity Are Associated With Differences in Negative
Mood During COVID-19 Lockdown. Front. Psychol. 2020,11, 588604. [CrossRef] [PubMed]
23.
Meyer, J.; McDowell, C.; Lansing, J.; Brower, C.; Smith, L.; Tully, M.; Herring, M. Changes in Physical Activity and Sedentary
Behavior in Response to COVID-19 and Their Associations with Mental Health in 3052 US Adults. Int. J. Environ. Res. Public.
Health 2020,17, 6469. [CrossRef] [PubMed]
24.
Yang, Y.; Koenigstorfer, J. Corrigendum to: Determinants of Physical Activity Maintenance during the Covid-19 Pandemic: A
Focus on Fitness Apps. Transl. Behav. Med. 2021, ibab145. [CrossRef]
25.
Rogers, N.T.; Waterlow, N.R.; Brindle, H.; Enria, L.; Eggo, R.M.; Lees, S.; Roberts, C.h. Behavioral Change Towards Reduced
Intensity Physical Activity Is Disproportionately Prevalent Among Adults with Serious Health Issues or Self-Perception of High
Risk During the UK COVID-19 Lockdown. Front. Public Health 2020,8, 575091. [CrossRef]
26. Smith, L.; Jacob, L.; Butler, L.; Schuch, F.; Barnett, Y.; Grabovac, I.; Veronese, N.; Caperchione, C.; Lopez-Sanchez, G.F.; Meyer, J.;
et al. Prevalence and Correlates of Physical Activity in a Sample of UK Adults Observing Social Distancing during the COVID-19
Pandemic. BMJ Open Sport Exerc. Med. 2020,6, e000850. [CrossRef]
27.
Cervero, R.; Duncan, M. Walking, Bicycling, and Urban Landscapes: Evidence from the San Francisco Bay Area. Am. J. Public
Health 2003,93, 1478–1483. [CrossRef]
28.
Humpel, N. Environmental Factors Associated with Adults’ Participation in Physical Activity A Review. Am. J. Prev. Med.
2002
,
22, 188–199. [CrossRef]
29.
Tanishita, M.; van Wee, B. Impact of Regional Population Density on Walking Behavior. Transp. Plan. Technol.
2017
,40, 661–678.
[CrossRef]
30.
Frank, L.D.; Schmid, T.L.; Sallis, J.F.; Chapman, J.; Saelens, B.E. Linking Objectively Measured Physical Activity with Objectively
Measured Urban Form. Am. J. Prev. Med. 2005,28, 117–125. [CrossRef]
Sustainability 2022,14, 3923 18 of 20
31.
Manaugh, K.; El-Geneidy, A. Validating Walkability Indices: How Do Different Households Respond to the Walkability of Their
Neighborhood? Transp. Res. Part Transp. Environ. 2011,16, 309–315. [CrossRef]
32.
Mertens, L.; Compernolle, S.; Gheysen, F.; Deforche, B.; Brug, J.; Mackenbach, J.D.; Lakerveld, J.; Oppert, J.-M.; Feuillet, T.; Glonti,
K.; et al. Perceived Environmental Correlates of Cycling for Transport among Adults in Five Regions of Europe: Environmental
Perceptions of Cycling. Obes. Rev. 2016,17, 53–61. [CrossRef] [PubMed]
33.
Harvey, C.; Aultman-Hall, L. Urban Streetscape Design and Crash Severity. Transp. Res. Rec. J. Transp. Res. Board
2015
,2500, 1–8.
[CrossRef]
34.
Brown, B.B.; Smith, K.R.; Hanson, H.; Fan, J.X.; Kowaleski-Jones, L.; Zick, C.D. Neighborhood Design for Walking and Biking.
Am. J. Prev. Med. 2013,44, 231–238. [CrossRef] [PubMed]
35. Southworth, M. Designing the Walkable City. J. Urban Plan. Dev. 2005,131, 246–257. [CrossRef]
36.
Braza, M.; Shoemaker, W.; Seeley, A. Neighborhood Design and Rates of Walking and Biking to Elementary School in 34 California
Communities. Am. J. Health Promot. 2004,19, 128–136. [CrossRef] [PubMed]
37.
Cerin, E.; Conway, T.L.; Saelens, B.E.; Frank, L.D.; Sallis, J.F. Cross-Validation of the Factorial Structure of the Neighborhood
Environment Walkability Scale (NEWS) and Its Abbreviated Form (NEWS-A). Int. J. Behav. Nutr. Phys. Act.
2009
,6, 32. [CrossRef]
38.
Berrigan, D.; Pickle, L.W.; Dill, J. Associations between Street Connectivity and Active Transportation. Int. J. Health Geogr.
2010
,
9, 20. [CrossRef]
39.
Kwon, M.; Remøy, H.; van den Bogaard, M. Influential Design Factors on Occupant Satisfaction with Indoor Environment in
Workplaces. Build. Environ. 2019,157, 356–365. [CrossRef]
40.
Christian, H.E.; Bull, F.C.; Middleton, N.J.; Knuiman, M.W.; Divitini, M.L.; Hooper, P.; Amarasinghe, A.; Giles-Corti, B. How
Important Is the Land Use Mix Measure in Understanding Walking Behaviour? Results from the RESIDE Study. Int. J. Behav.
Nutr. Phys. Act. 2011,8, 55. [CrossRef]
41. Wei, Y.; Xiao, W.; Wen, M.; Wei, R. Walkability, Land Use and Physical Activity. Sustainability 2016,8, 65. [CrossRef]
42.
Ewing, R.; Handy, S. Measuring the Unmeasurable: Urban Design Qualities Related to Walkability. J. Urban Des.
2009
,14, 65–84.
[CrossRef]
43.
Lynch, K. City Sense and City Design: Writings and Projects of Kevin Lynch; Banerjee, T., Ed.; MIT Press: Cambridge, MA, USA, 2006;
ISBN 978-0-262-62095-6.
44.
Millington, C.; Ward Thompson, C.; Rowe, D.; Aspinall, P.; Fitzsimons, C.; Nelson, N.; Mutrie, N. Development of the Scottish
Walkability Assessment Tool (SWAT). Health Place 2009,15, 474–481. [CrossRef] [PubMed]
45.
Cerin, E.; Nathan, A.; van Cauwenberg, J.; Barnett, D.W.; Barnett, A. The Neighbourhood Physical Environment and Active
Travel in Older Adults: A Systematic Review and Meta-Analysis. Int. J. Behav. Nutr. Phys. Act. 2017,14, 15. [CrossRef]
46.
Bowling, A.; Gabriel, Z.; Dykes, J.; Dowding, L.M.; Evans, O.; Fleissig, A.; Banister, D.; Sutton, S. Let’s Ask Them: A National
Survey of Definitions of Quality of Life and Its Enhancement among People Aged 65 and Over. Int. J. Aging Hum. Dev.
2003
,56,
269–306. [CrossRef]
47.
De Vries, S.; Verheij, R.A.; Groenewegen, P.P.; Spreeuwenberg, P. Natural Environments—Healthy Environments? An Exploratory
Analysis of the Relationship between Greenspace and Health. Environ. Plan. Econ. Space 2003,35, 1717–1731. [CrossRef]
48.
Nezlek, J.B.; Leary, M.R. Individual Differences in Self-Presentational Motives in Daily Social Interaction. Pers. Soc. Psychol. Bull.
2002,28, 211–223. [CrossRef]
49.
French, D.P.; Darker, C.D.; Eves, F.F.; Sniehotta, F.F. The Systematic Development of a Brief Intervention to Increase Walking in the
General Public Using an “Extended” Theory of Planned Behavior. J. Phys. Act. Health 2013,10, 940–948. [CrossRef]
50.
Brown, B.B.; Werner, C.M.; Smith, K.R.; Tribby, C.P.; Miller, H.J. Physical Activity Mediates the Relationship between Perceived
Crime Safety and Obesity. Prev. Med. 2014,66, 140–144. [CrossRef]
51.
Fish, J.S.; Ettner, S.; Ang, A.; Brown, A.F. Association of Perceived Neighborhood Safety on Body Mass Index. Am. J. Public Health
2010,100, 2296–2303. [CrossRef]
52.
Deutsch, K.; Yoon, S.Y.; Goulias, K. Modeling Travel Behavior and Sense of Place Using a Structural Equation Model. J. Transp.
Geogr. 2013,28, 155–163. [CrossRef]
53.
Etminani-Ghasrodashti, R.; Ardeshiri, M. Modeling Travel Behavior by the Structural Relationships between Lifestyle, Built
Environment and Non-Working Trips. Transp. Res. Part Policy Pract. 2015,78, 506–518. [CrossRef]
54.
Hurtubia, R.; Atasoy, B.; Glerum, A.; Curchod, A.; Bierlaire, M. Considering Latent Attitudes in Mode Choice: The Case of
Switzerland. In Proceedings of the 12th WCTR Conference, Lisbon, Portugal, 11–15 July 2010; pp. 11–15.
55.
Zarabi, Z.; Lord, S. Toward More Sustainable Behavior: A Systematic Review of the Impacts of Involuntary Workplace Relocation
on Travel Mode Choice. J. Plan. Lit. 2019,34, 38–58. [CrossRef]
56.
Ball, K. Traversing Myths and Mountains: Addressing Socioeconomic Inequities in the Promotion of Nutrition and Physical
Activity Behaviours. Int. J. Behav. Nutr. Phys. Act. 2015,12, 142. [CrossRef] [PubMed]
57.
Ball, K.; Abbott, G.; Wilson, M.; Chisholm, M.; Sahlqvist, S. How to Get a Nation Walking: Reach, Retention, Participant
Characteristics and Program Implications of Heart Foundation Walking, a Nationwide Australian Community-Based Walking
Program. Int. J. Behav. Nutr. Phys. Act. 2017,14, 161. [CrossRef]
58.
Kassavou, A.; Turner, A.; French, D.P. Do Interventions to Promote Walking in Groups Increase Physical Activity? A Meta-
Analysis. Int. J. Behav. Nutr. Phys. Act. 2013,10, 18. [CrossRef]
Sustainability 2022,14, 3923 19 of 20
59.
Hanson, S.; Jones, A. Is There Evidence That Walking Groups Have Health Benefits? A Systematic Review and Meta-Analysis. Br.
J. Sports Med. 2015,49, 710–715. [CrossRef]
60.
Betancourt, T.S.; McBain, R.; Newnham, E.A.; Akinsulure-Smith, A.M.; Brennan, R.T.; Weisz, J.R.; Hansen, N.B. A Behavioral
Intervention for War-Affected Youth in Sierra Leone: A Randomized Controlled Trial. J. Am. Acad. Child Adolesc. Psychiatry
2014
,
53, 1288–1297. [CrossRef]
61.
Phillips, L.T.; Slepian, M.L.; Hughes, B.L. Perceiving Groups: The People Perception of Diversity and Hierarchy. J. Pers. Soc.
Psychol. 2018,114, 766–785. [CrossRef]
62.
Moudon, A.V.; Lee, C.; Cheadle, A.D.; Garvin, C.; Johnson, D.B.; Schmid, T.L.; Weathers, R.D. Attributes of Environments
Supporting Walking. Am. J. Health Promot. 2007,21, 448–459. [CrossRef]
63.
Owen, N.; Humpel, N.; Leslie, E.; Bauman, A.; Sallis, J.F. Understanding Environmental Influences on Walking. Am. J. Prev. Med.
2004,27, 67–76. [CrossRef] [PubMed]
64.
López-Bueno, R.; Calatayud, J.; Andersen, L.L.; Balsalobre-Fernández, C.; Casaña, J.; Casajús, J.A.; Smith, L.; López-Sánchez, G.F.
Immediate Impact of the COVID-19 Confinement on Physical Activity Levels in Spanish Adults. Sustainability
2020
,12, 5708.
[CrossRef]
65.
Stanton, R.; To, Q.G.; Khalesi, S.; Williams, S.L.; Alley, S.J.; Thwaite, T.L.; Fenning, A.S.; Vandelanotte, C. Depression, Anxiety and
Stress during COVID-19: Associations with Changes in Physical Activity, Sleep, Tobacco and Alcohol Use in Australian Adults.
Int. J. Environ. Res. Public. Health 2020,17, 4065. [CrossRef] [PubMed]
66.
Maugeri, G.; Castrogiovanni, P.; Battaglia, G.; Pippi, R.; D’Agata, V.; Palma, A.; Di Rosa, M.; Musumeci, G. The Impact of Physical
Activity on Psychological Health during Covid-19 Pandemic in Italy. Heliyon 2020,6, e04315. [CrossRef] [PubMed]
67.
Cheval, B.; Sivaramakrishnan, H.; Maltagliati, S.; Fessler, L.; Forestier, C.; Sarrazin, P.; Orsholits, D.; Chalabaev, A.; Sander, D.;
Ntoumanis, N.; et al. Relationships between Changes in Self-Reported Physical Activity, Sedentary Behaviour and Health during
the Coronavirus (COVID-19) Pandemic in France and Switzerland. J. Sports Sci. 2021,39, 699–704. [CrossRef] [PubMed]
68.
Xu, R.; Lin, Y.; Zhang, B. Relationship among Sleep Quality, Depressed Mood, and Perceived Social Support in Older Adults: A
Longitudinal Study. J. Pac. Rim Psychol. 2021,15, 183449092110526. [CrossRef]
69.
Orgilés, M.; Morales, A.; Delvecchio, E.; Francisco, R.; Mazzeschi, C.; Pedro, M.; Espada, J.P. Coping Behaviors and Psychological
Disturbances in Youth Affected by the COVID-19 Health Crisis. Front. Psychol. 2021,12, 565657. [CrossRef]
70.
Ammar, A.; Brach, M.; Trabelsi, K.; Chtourou, H.; Boukhris, O.; Masmoudi, L.; Bouaziz, B.; Bentlage, E.; How, D.; Ahmed, M.; et al.
Effects of COVID-19 Home Confinement on Eating Behaviour and Physical Activity: Results of the ECLB-COVID19 International
Online Survey. Nutrients 2020,12, 1583. [CrossRef]
71.
Brownson, R.C.; Baker, E.A.; Housemann, R.A.; Brennan, L.K.; Bacak, S.J. Environmental and Policy Determinants of Physical
Activity in the United States. Am. J. Public Health 2001,91, 1995–2003. [CrossRef]
72.
Clark, A.; Scott, D. Understanding the Impact of the Modifiable Areal Unit Problem on the Relationship between Active Travel
and the Built Environment. Urban Stud. 2014,51, 284–299. [CrossRef]
73.
Mason, P.; Kearns, A.; Livingston, M. “Safe Going”: The Influence of Crime Rates and Perceived Crime and Safety on Walking in
Deprived Neighbourhoods. Soc. Sci. Med. 2013,91, 15–24. [CrossRef] [PubMed]
74.
Owen, N.; Cerin, E.; Leslie, E.; duToit, L.; Coffee, N.; Frank, L.D.; Bauman, A.E.; Hugo, G.; Saelens, B.E.; Sallis, J.F. Neighborhood
Walkability and the Walking Behavior of Australian Adults. Am. J. Prev. Med. 2007,33, 387–395. [CrossRef] [PubMed]
75.
Gomes, G.A.; Reis, R.S.; Parra, D.C.; Ribeiro, I.; Hino, A.A.; Hallal, P.C.; Malta, D.C.; Brownson, R.C. Walking for Leisure among
Adults from Three Brazilian Cities and Its Association with Perceived Environment Attributes and Personal Factors. Int. J. Behav.
Nutr. Phys. Act. 2011,8, 111. [CrossRef] [PubMed]
76.
Özbil, A.; Peponis, J. The Effects of Urban Form on Walking to Transit. In Proceedings of the 8th International Space Syntax
Symposium, Santiago de Chile, Chile, 3–6 January 2012.
77.
Giehl, M.W.C.; Hallal, P.C.; Corseuil, C.W.; Schneider, I.J.C.; d’Orsi, E. Built Environment and Walking Behavior Among Brazilian
Older Adults: A Population-Based Study. J. Phys. Act. Health 2016,13, 617–624. [CrossRef] [PubMed]
78.
Masoumi, H.E. Modeling the Travel Behavior Impacts of Micro-Scale Land Use and Socio-Economic Factors. TeMA J. Land Use
Mobil. Environ. 2013,6, 235–250. [CrossRef]
79.
Ahmadipour, F.; Mamdoohi, A.R.; Wulf-Holger, A. Impact of Built Environment on Walking in the Case of Tehran, Iran. J. Transp.
Health 2021,22, 101083. [CrossRef]
80.
Ermagun, A.; Levinson, D. Public Transit, Active Travel, and the Journey to School: A Cross-Nested Logit Analysis. Transp.
Transp. Sci. 2017,13, 24–37. [CrossRef]
81. Ermagun, A.; Samimi, A. Promoting Active Transportation Modes in School Trips. Transp. Policy 2015,37, 203–211. [CrossRef]
82.
Hatamzadeh, Y.; Hosseinzadeh, A. Toward a Deeper Understanding of Elderly Walking for Transport: An Analysis across
Genders in a Case Study of Iran. J. Transp. Health 2020,19, 100949. [CrossRef]
83.
World Health Organization. Personal Interventions and Risk Communication on Air Pollution: Summary Report of WHO Expert
Consultation; World Health Organization: Geneva, Switzerland, 2020; ISBN 978-92-4-000027-8.
84.
Vale, D.S.; Pereira, M. Influence on Pedestrian Commuting Behavior of the Built Environment Surrounding Destinations: A
Structural Equations Modeling Approach. Int. J. Sustain. Transp. 2016,10, 730–741. [CrossRef]
85.
Chan, E.T.H.; Schwanen, T.; Banister, D. The Role of Perceived Environment, Neighbourhood Characteristics, and Attitudes in
Walking Behaviour: Evidence from a Rapidly Developing City in China. Transportation 2021,48, 431–454. [CrossRef]
Sustainability 2022,14, 3923 20 of 20
86.
Rhodes, R.E.; Liu, S.; Lithopoulos, A.; Zhang, C.; Garcia-Barrera, M.A. Correlates of Perceived Physical Activity Transitions
during the COVID-19 Pandemic among Canadian Adults. Appl. Psychol. Health Well-Being
2020
,12, 1157–1182. [CrossRef]
[PubMed]
87.
Teran-Escobar, C.; Forestier, C.; Ginoux, C.; Isoard-Gautheur, S.; Sarrazin, P.; Clavel, A.; Chalabaev, A. Individual, Sociodemo-
graphic, and Environmental Factors Related to Physical Activity During the Spring 2020 COVID-19 Lockdown. Front. Psychol.
2021,12, 643109. [CrossRef] [PubMed]
88.
Hargreaves, E.A.; Lee, C.; Jenkins, M.; Calverley, J.R.; Hodge, K.; Houge Mackenzie, S. Changes in Physical Activity Pre-, During
and Post-Lockdown COVID-19 Restrictions in New Zealand and the Explanatory Role of Daily Hassles. Front. Psychol.
2021
,
12, 642954. [CrossRef]
89.
Constandt, B.; Thibaut, E.; De Bosscher, V.; Scheerder, J.; Ricour, M.; Willem, A. Exercising in Times of Lockdown: An Analysis of
the Impact of COVID-19 on Levels and Patterns of Exercise among Adults in Belgium. Int. J. Environ. Res. Public. Health
2020
,
17, 4144. [CrossRef]
90.
Kaur-Gill, S.; Qin-Liang, Y.; Hassan, S. Negotiating Mental Health During the COVID-19 Pandemic: Performing Migrant Domestic
Work in Contentious Conditions. Am. Behav. Sci. 2021,65, 1406–1425. [CrossRef]
91.
Mehriar, M.; Masoumi, H.; Aslam, A.B.; Gillani, S.M. The Neighborhood Effect on Keeping Non-Commuting Journeys within
Compact and Sprawled Districts. Land 2021,10, 1245. [CrossRef]
92.
Winters, M.; Buehler, R.; Götschi, T. Policies to Promote Active Travel: Evidence from Reviews of the Literature. Curr. Environ.
Health Rep. 2017,4, 278–285. [CrossRef]
93.
Muñoz, B.; Monzon, A.; Ricardo, A. The Increasing Role of Latent Variables in Modelling Bicycle Mode Choice. In Transport
Review; Routledge: Washington, WA, USA, 2016; Volume 36, pp. 737–771.
94.
Wang, H.; Feng, L.; Zhang, Y.; Zhang, F.; Fu, J.; Wang, M.; Wu, D.; Feng, Q.; Liu, X.; Fan, C.; et al. Changes in Chinese Adults’
Physical Activity Behavior and Determinants before and during the COVID-19 Pandemic. J. Clin. Med.
2021
,10, 3069. [CrossRef]
95.
Lear, S.A.; Hu, W.; Rangarajan, S.; Gasevic, D.; Leong, D.; Iqbal, R.; Casanova, A.; Swaminathan, S.; Anjana, R.M.; Kumar, R.; et al.
The Effect of Physical Activity on Mortality and Cardiovascular Disease in 130 000 People from 17 High-Income, Middle-Income,
and Low-Income Countries: The PURE Study. Lancet 2017,390, 2643–2654. [CrossRef]
96.
Boakye-Dankwa, E.; Barnett, A.; Pachana, N.A.; Turrell, G.; Cerin, E. Associations Between Latent Classes of Perceived Neighbor-
hood Destination Accessibility and Walking Behaviors in Older Adults of a Low-Density and a High-Density City. J. Aging Phys.
Act. 2019,27, 553–564. [CrossRef] [PubMed]
97.
Newman, P.; Kenworthy, J.R. Cities and Automobile Dependence: A Sourcebook; Gower Technical: Aldershot, UK; Brookfield, VT,
USA, 1989; ISBN 978-0-566-07040-2.
98.
Holden, E.; Norland, I.T. Three Challenges for the Compact City as a Sustainable Urban Form: Household Consumption of
Energy and Transport in Eight Residential Areas in the Greater Oslo Region. Urban Stud. 2005,42, 2145–2166. [CrossRef]
99.
De Moor, D. Walking Works: Making the Case to Encourage Greater Uptake of Walking as a Physical Activity and Recognize the Value and
Benefits of Walking for Health; The Ramblers: London, UK, 2013.
100.
Gebel, K.; Bauman, A.; Owen, N. Correlates of Non-Concordance between Perceived and Objective Measures of Walkability. Ann.
Behav. Med. 2009,37, 228–238. [CrossRef] [PubMed]
101.
Gebel, K.; Bauman, A.E.; Sugiyama, T.; Owen, N. Mismatch between Perceived and Objectively Assessed Neighborhood
Walkability Attributes: Prospective Relationships with Walking and Weight Gain. Health Place
2011
,17, 519–524. [CrossRef]
[PubMed]
102.
Galea, M.N.; Bray, S.R. Predicting Walking Intentions and Exercise in Individuals with Intermittent Claudication: An Application
of the Theory of Planned Behavior. Rehabil. Psychol. 2006,51, 299–305. [CrossRef]
103.
Sun, G.; Acheampong, R.; Lin, H.; Pun, V. Understanding Walking Behavior among University Students Using Theory of Planned
Behavior. Int. J. Environ. Res. Public. Health 2015,12, 13794–13806. [CrossRef]
104.
Cummins, S.; Curtis, S.; Diez-Roux, A.V.; Macintyre, S. Understanding and Representing ‘Place’ in Health Research: A Relational
Approach. Soc. Sci. Med. 2007,65, 1825–1838. [CrossRef]
105.
Hunter, R.F.; Garcia, L.; de Sa, T.H.; Zapata-Diomedi, B.; Millett, C.; Woodcock, J.; Pentland, A.; Moro, E. Effect of COVID-19
Response Policies on Walking Behavior in US Cities. Nat. Commun. 2021,12, 3652. [CrossRef]
106.
EARG Collaborators; Abouzid, M.; El-Sherif, D.M.; Eltewacy, N.K.; Dahman, N.B.H.; Okasha, S.A.; Ghozy, S.; Islam, S.M.S.
Influence of COVID-19 on Lifestyle Behaviors in the Middle East and North Africa Region: A Survey of 5896 Individuals. J.
Transl. Med. 2021,19, 129. [CrossRef]
107.
Zafri, N.M.; Khan, A.; Jamal, S.; Alam, B.M. Impacts of the COVID-19 Pandemic on Active Travel Mode Choice in Bangladesh: A
Study from the Perspective of Sustainability and New Normal Situation. Sustainability 2021,13, 6975. [CrossRef]
108.
Habib, M.A.; Anik, M.A.H. Impacts of COVID-19 on Transport Modes and Mobility Behavior: Analysis of Public Discourse in
Twitter. Transp. Res. Rec. J. Transp. Res. Board 2021, 036119812110299. [CrossRef]
