Transport Policy 108 (2021) 21–33

Contents lists available at ScienceDirect

Transport Policy
journal homepage: www.elsevier.com/locate/tranpol

Measuring changes in travel behavior pattern due to COVID-19 in a

developing country: A case study of Pakistan
Muhammad Abdullah a, *, Nazam Ali a, Syed Arif Hussain b, Atif Bilal Aslam c,
Muhammad Ashraf Javid d
a
Department of Civil Engineering, School of Engineering, University of Management and Technology, Lahore, 54770, Pakistan
b
Faculty of Civil and Environmental Engineering, Graduate School of Science & Engineering, Saitama University, Saitama, 338-8570, Japan
c
Department of City and Regional Planning, University of Engineering and Technology, Lahore, Punjab, 54890, Pakistan
d
Department of Civil and Environmental Engineering, College of Engineering and Architecture, University of Nizwa, Birkat-al-Mouz, 616, Nizwa, Oman

A R T I C L E I N F O A B S T R A C T

Keywords: Travel behavior has been affected around the world since the eruption of corona virus disease (COVID-19).
COVID-19 Several industries including transportation industry have been hard hit by the pandemic. As the virus is trans­
SARS-CoV-2 mitted through close contact with infected people, number of outdoor trips has reduced causing roads and public
Pandemic
transport to be less crowded than before. In order to develop transport-related policies for the post COVID-19
Travel behavior
Mode choice
world, it is necessary to explore how the pandemic has affected the travel behavior pattern. This study
Pakistan explored the influence of the COVID-19 pandemic on travel pattern and mode preferences in Pakistan using a
questionnaire survey. The results showed significant shift in primary traveling purpose from work and studying
to shopping during the pandemic. Number of trips performed for non-commuting purposes were also signifi­
cantly different before and during the pandemic. A significant modal shift from motorbike to non-motorized
modes of travel was found for distances less than 5 km. For longer distances, people shifted from public trans­
port to private car. These findings suggest that past policies regarding different modes may be revisited in the
post COVID-19 world. The statistical tests performed on the factors affecting mode choices indicated that the
respondents put more priority on pandemic-related items such as infection concern, social distance, hand san­
itizers’ availability, and cleanliness, etc., during the pandemic. The findings of this study will certainly help in
shaping up the policies for the post COVID-19 world especially in the developing countries.

1. Introduction coronavirus, or HCoV can live on metal, plastic, or glass (inanimate)

surfaces for about 9 days at 30 ◦ C and the duration of persistence reduces
The eruption of the 2019 novel coronavirus causing COVID-19 has with increase in temperature (Kampf et al., 2020). Van Doremalen et al.,
influenced travel behavior around the world. The first case of this virus 2020 and Wu et al. (2020) suggested that aerosol and surface stability of
was recorded on December 12, 2019 in Hebei, Wuhan, China. It was COVID-19 is similar to MERS, SARS, and HCoV. However, these viruses
then labeled a global pandemic on March 11, 2020 (McCloskey and can be efficiently inactivated if the surfaces are disinfected and their
Heymann, 2020). A total number of 84.47 million people had been spread can be limited by exercising basic hygiene guidelines issued by
infected with 1.84 million (approximately 2.18%) fatalities around the WHO such as; wearing a face mask, avoiding crowded places, and
world till January 06, 2021. The first case of COVID-19 infection in washing hands (WHO, 2020). Maintaining a social distance has been
Pakistan was recorded on February 26, 2020 leading to 725,602 in­ found to be an efficient strategy until a vaccine is developed.
fections and 15,501 (2.13%) fatalities on April 13, 2021 (WHO, 2020). After the outbreak of the COVID-19, the governments around the
The disease caused by the virus is very contagious and spreads globe took unprecedented actions to contain the virus. Many countries;
through close contact with infected persons. In addition, epidemiologists like France, Italy, UK, Iran, India, Saudi Arabia and Thailand imple­
reported that COVID-19 virus can live on hard surfaces from few hours mented national lockdowns, limiting all non-essential traveling (de Haas
to several days. Human corona viruses such as, SARS coronavirus, MERS et al. 2020). Likewise, Pakistan initiated nation-wide lockdown on April

* Corresponding author.
E-mail address: [email protected] (M. Abdullah).

https://doi.org/10.1016/j.tranpol.2021.04.023
Received 10 January 2021; Received in revised form 16 April 2021; Accepted 28 April 2021
Available online 5 May 2021
0967-070X/© 2021 Elsevier Ltd. All rights reserved.
M. Abdullah et al. Transport Policy 108 (2021) 21–33

01, 2020, and extended it twice until May 09, 2020. The lockdown was established an association between the mobility pattern and the
eased in phases but some of the areas in mega cities with severe clusters COVID-19 spread. The authors of the emerging studies on the topic also
remained under smart lockdown. categorized their work in the context of developed and developing
Travel behavior has been affected around the world since the erup­ countries (such as Mogaji, 2020) to look at the similarities and the
tion of the COVID-19 disease. It is more than a health crisis because variations of the results across different socioeconomic contexts.
several industries including transportation industry have been hard hit The first direct influence of the COVID-19 pandemic on the travel
by the pandemic. As the virus is transmitted through close contact with pattern was the implementation of mobility restrictions, which has
infected people, number of outdoor trips have reduced causing roads resulted in the disruption of livelihoods and social life of the people
and public transport to be less crowded than before. Besides other socio- (Mogaji, 2020; Musselwhite et al., 2020; SLOCAT, 2020). Musselwhite
economic impacts of this outbreak, this unique situation has compelled et al. (2020) argue that the multifaceted dimensions of COVID-19 are
people to alter their daily lives drastically within a span of days and affecting the travel pattern across the globe and that the associated
weeks (Musselwhite et al., 2020). Work and travel, the basic facets of changes may be long lasting in the post-COVID-19 world. The economic
daily lives, have changed radically due to this outbreak. In developing constraints posed by the COVID-19 might force the people to reduce the
countries, like Pakistan, public transport (buses, wagons, metros, trains car usage frequency in favor of walking, cycling or cheaper public
etc.) carries millions of people, often transporting passengers beyond transport modes (Litman, 2020). However, predicting the future
their designated capacity with little or no consideration for hygiene and changes in the travel pattern is not that straight forward because people
personal space. Moreover, maintaining a social distance (approx. 2 m) in might switch from public transport to private cars due to safety and
public transportation seems a daunting task, which may affect mode health woes (Stokell, 2020). It is more pertinent to those places which
choice behavior. lack cycling and walking infrastructure, particularly in the developing
To develop transport-related policies for the post COVID-19 world, it world. The COVID-19 consequences have also brought the importance of
is necessary to explore how the pandemic has affected the travel walkable and car-independent accessible neighborhoods in the lime­
behavior pattern. This study aims to explore how COVID-19 affected the light. The overall travel trips, mainly the passenger transport trips have
travel behavior in Punjab, the most populated province of Pakistan. It been reduced globally (EU, 2020; ITE, 2020; TUMI, 2020; UITP, 2020),
focuses on the changes in primary traveling purpose, mode choice and but it is argued that this change will be temporary, and the
factors influencing mode choice before as well as during the COVID-19 pre-COVID-19 travel pattern will return as the direct effects of
pandemic. It also charts out new guidelines and policies for public and COVID-19 on the economy will fade away (Litman, 2020).
private stakeholders to address the mobility needs of individuals for West et al. (2020) identified human behavior as the central element
current as well as future pandemics. The findings of this study will for the transmission of the COVID-19 virus. Wilder-Smith et al. (2020)
certainly help in shaping up the policies for post the COVID-19 world, argued on the important role of the overall societal response and
especially in the developing countries. top-down measurements to contain the coronavirus disease. A growing
The rest of the paper is organized as follows: Next section presents number of emerging studies are also providing an evidence on the in­
relevant studies. Section 3 presents the methodology. Section 4 presents fluence of travel behavior on the transmission of COVID-19 globally.
the results and discussion. Important policy implications are presented Kraemer et al. (2020) found a correlation between the human mobility
in Section 5. Finally, Section 6 presents conclusions and recommenda­ pattern and COVID-19 spread in China. However, after the imposition of
tions for future works. control measures and travel restrictions, the analysis of the travel data
showed an insignificant correlation between the two variables under
2. Literature review consideration. Peeri et al. (2020) also presented the similar results and
found the massive movements of people to and from Wuhan during the
The spread of infectious diseases has been found to affect travel Chinese New Year celebrations as one of the main reasons behind the
behavior and vice versa. Mandatory lockdowns and the fear of getting COVID-19 propagation. Zhang et al. (2020) also confirmed the positive
infected generally reduce outdoor trips (Cahyanto et al., 2016; Kim role of high-speed trains and air flights in spreading the COVID-19 cases
et al., 2017). During the MERS outbreak in South Korea, trips to affected across Chinese cities.
and other areas were reduced by about 14% and 9%, respectively. In Various travel modes have their merits and demerits over each other
addition, about 10% reduction in public transport trips was observed with respect to the risks associated with COVID-19 spread. Public
during the same outbreak. On the one hand, infectious diseases affect the transport and active traveling modes have been advocated much in the
travel behavior, while on the other hand, travel behavior can help recent decades as means of sustainable mobility. However, social
predict the spread of infectious diseases. For example, a review of distancing, which is an agreed upon precautionary measure against
several studies suggested that highways, railways, and air transport COVID-19, is in conflict with the principle of public transportation
propagate the spread of influenza and coronaviruses (Zhang et al., 2011; (Musselwhite et al., 2020). For instance, an association has been re­
Browne et al., 2016). Hence, appropriate travel restrictions and social ported between acute respiratory infection (ARI) and the use of buses
distancing measures can limit the spread of infectious diseases (Milne and tram lines (Troko et al., 2011). Public transportation modes offer a
et al., 2008; Otsuki and Nishiura, 2016). Ibuka et al. (2010) also iden­ closed environment for the riders to remain either seated or standing for
tified that the individual behaviors and perceptions are also significant a considerable time nearby each other., thus they greatly increase the
predictors of adopted travel patterns such as staying at home and limited risks of infectious diseases spread (Edelson and Phypers, 2011). How­
interaction during infectious diseases outbreak. Since there is a two-way ever, Musselwhite et al. (2020), by referring the works of Williams et al.
relationship between spread of infectious diseases and travel behavior, (2010) and Cooley et al. (2011), do not advocate the idea of suspension
the current ongoing phenomenon of COVID-19 outbreak is also being of mass public transport vehicles during the pandemic as a counter­
studied through two different lines of investigations with respect to its measure to the containment of coronavirus related infectious diseases,
relationship with travel behavior. One strand of studies attempts to as to them, household exposure poses a much greater risk for the spread
explore the effects of the COVID-19 pandemic on travel pattern (Litman, of such infectious diseases. Also, Peeri et al. (2020) reported inadequate
2020; Bucsky, 2020; Ebenso & Otu 2020; de Haas et al., 2020) while the risk assessment and limited reporting of virus cases in China as the main
other strand of studies investigates the role of travel pattern in pre­ reason behind the rapid spread of COVID-19. Though, they also reported
dicting the COVID-19 spread (such as Kraemer et al., 2020). Many recent on the significant role of increased accessibility of Wuhan and associated
studies of former type have found out significant changes in the travel travel pattern observed by the Chinese in spreading the virus. None­
behavior and mode preferences of the people across the globe due to the theless, public transportation can still be safely used during the
COVID-19 pandemic, while the latter strand of studies has also COVID-19 pandemic by adhering to precautionary measures by its users

22
M. Abdullah et al. Transport Policy 108 (2021) 21–33

(APTA, 2020), although an entirely risk-free environment for traveling recommendations were incorporated to ensure the clarity and under­
will not be possible (EU, 2020). The precautionary measures with standing of each statement to the respondents. Face-to-face interviews
respect to public transport use, which are being highly prioritized are were not possible due to the social distancing requirements and the
physical distancing, avoiding congestion and crowding, disinfection of partial lockdown in the country. Therefore, an online questionnaire was
public transport vehicles and transit hubs, and availability of sanitizers distributed to the target population through emails, social media web­
in all public transport vehicles and hubs (EU, 2020; Transit Center, sites and personal contacts. The responses were received from Lahore,
2020). Evidence emerging from Japan and France suggests the effec­ Faisalabad, and Rawalpindi, three major cities of the most populated
tiveness of the adopted precautionary measures in relation to public province of Pakistan, Punjab. These three cities are the most populated
transit use during the pandemic, as none of the spotted COVID-19 cities in the Punjab province and have similar socio-economic charac­
clusters could be traced to the public transits in both countries (O’Sul­ teristics. The survey was run for a period of about three weeks from May
livan, 2020). 09 to 31, 2020. Six hundred and seventy-one (671) responses were
Whereas the active traveling modes of cycling and walking have received during this period. It should be noted that the lockdown was
been associated with many health-related advantages (Hamer and lifted on May 09 in Pakistan and only few areas were under smart
Chider, 2008; Reinhardt-Rutland, 2011). Recently, cycling and walking lockdown after May 09. The timeline of COVID-19, cumulative number
have been regarded as socially compatible with the social distancing of cases and associated deaths in Pakistan (OurWorldinData, 2020) are
protocols by many researchers in an open letter to the UK government shown in Fig. 1. It also shows the data provided by Punjab Safe Cities
(Woodcock et al., 2020). Extending the pavements and bicycle paths for Authority (PSCA) about changes in traffic levels at various major roads
active traveling has already been thought out as one of the urban in Lahore, Punjab, Pakistan. The term lockdown means the complete
mobility way-forward options in many of the member states of the Eu­ nationwide lockdown implemented by the government, whereas smart
ropean Union (EU, 2020). Many local level initiatives taken across the lockdown means partial lockdown which was implemented in certain
globe also support the adoption of walking and cycling interventions areas containing clusters of the COVID-19 cases after the complete
more often during the COVID-19 pandemic (PBIS, 2020). Most of the lockdown had been lifted.
developed world countries and municipalities (examples are UK,
Madrid, Lisbon, Bordeaux, Athens, and many more) have put in place 3.2. Survey instrument
the transport related programs promoting the use of e-scooters, cycling,
and walking in response to the COVID-19 pandemic (POLIS, 2020). The questionnaire had three distinct sections: (1) socio-economic
The COVID-19 pandemic is a global event which has affected and demographic characteristics (SEDs), (2) travel pattern before as
numerous aspects of daily lives. There is a lack of enough empirical well as during the COVID-19 pandemic, and (3) the factors influencing
evidence on how this pandemic has affected travel behavior in devel­ mode preferences before as well as during the COVID-19 pandemic.
oping countries. This research study fills the gap by exploring how SEDs consisted of age, age, monthly household income, education level,
COVID-19 pandemic is influencing travel behaviours and mode prefer­ cycle ownership, motorbike ownership, car ownership, and employment
ences in developing countries such as Pakistan. It also aims at outlining status. Section 2 contained questions about primary traveling purpose,
new transport policies considering the travel behavior changes due to the main reason for which people travel, before as well as during the
the pandemic. It offers new insights and practical implications for both COVID-19 pandemic. It also contained questions about traveling fre­
public and private policy makers and stakeholders to navigate through quency for non-commuting trips and preferred mode for short as well as
these precarious times and chart new pathways for individual travel long distances before and during the COVID-19 pandemic. Section 3
behaviours. consisted of 5-point Likert type items representing factors that may in­
fluence mode preferences before and during the COVID-19 pandemic.
3. Methods Since the virus spreads primarily through close contact with infected
people or by getting in contact with an infected surface, it is hypothe­
3.1. Survey design and sampling strategy sized that people will pay more attention to pandemic-related items
while choosing a transport mode. For example, comfort and convenience
The questionnaire was prepared carefully to measure the travel might not be as important as infection concern and maintaining social
pattern both before and during the COVID-19 pandemic. The objectives distance during the pandemic. Therefore, factors, which may affect
of the study and the instructions to fill the questionnaire were provided mode choice during the pandemic, were identified and respondents
to the respondents to obtain reliable responses. The questionnaire was were asked to place a priority on each factor while choosing a transport
pre-tested by the experts belonging to the relevant research area and the mode.

Fig. 1. COVID-19 timeline, cumulative cases and deaths in Pakistan, and traffic levels at various major roads in Lahore, Punjab, Pakistan.

23
M. Abdullah et al. Transport Policy 108 (2021) 21–33

3.3. Analysis methods Table 1

SEDs of the respondents.
Since nonparametric tests require fewer assumptions as compared to Items Description %
parametric equivalents, can be conducted on ordinal, and are relatively
Gender Male 73.3
easy to comprehend (Gibbons and Chakraborti, 2014), therefore, they Female 26.7
were used in this study. Nonparametric tests may have lower power as
Age 18–30 65.9
compared to their counterparts, however, the power loss will be rela­ 31–45 30.0
tively small if the data follows a normal distribution and meets all other 46–60 3.3
assumptions (Kitchen, 2009). It is suggested that nonparametric tests be Above 60 0.9
used instead of parametric tests if there is no experimental evidence Education level High schooling .4
about the distribution of the errors. The data collected through Section 2 College 2.5
and 3 of the questionnaire consisted of repeated measurements, which Bachelors 44.9
require special considerations during the statistical analysis. SPSS v. 20 Masters and above 52.2

was used for all the statistical analyses. Occupation Student 37.0
Business 4.0
Government employee 24.9
3.3.1. Statistical analyses on repeated measurements
Private employee 28.8
McNemar-Bowker test was applied on the repeated measurements of Others 5.4
nominal data e.g., primary traveling purpose before and during the
Income level (PKR) Less than 25,000 5.8
pandemic. It is a nonparametric test with more than 2 levels for each 25,000–50,000 23.0
nominal variable which can be presented in a k × k contingency matrix. 50,001–100,000 28.2
Moreover, McNemar’s tests were carried out as post-hoc tests to deter­ More than 100,000 28.2
mine the levels of nominal variables which differed significantly. The Prefer not to say 14.9

McNemar test is used to find out if there are differences on a binary Car Ownership Yes 42.9
dependent variable between two paired groups. In order to reduce type I No 57.1
errors while carrying out multiple hypothesis testing, Bonferroni Motorcycle Ownership Yes 51.1
correction was used, which is calculated as α/n, where α and n represent No 48.9
the significance level and the number of tests to be conducted, Bicycle Ownership Yes 12.8
respectively. No 87.2
To compare the ordinal repeated measurements under two different
conditions, such as number of non-commuting trips made before and
the share of working women is quite less in the population due to cul­
during the COVID-19 pandemic, a non-parametric alternative of the
tural and religious constraints. Hence, the share of female respondents
paired t-test known as Wilcoxon signed rank test was used.
makes a good representation of the women population whose travel
behavior was likely impacted by the pandemic. Similarly, the higher
3.3.2. Statistical analyses on independent measurements
percentage of young and highly educated respondents may imply that
The Mann-Whitney U test can be used to compare differences be­
young and educated people are more active on social media as compared
tween two independent groups. It is a non-parametric alternative to the
to the less educated and older respondents i.e., 46–60 years old (3.3%)
two-sample t-test and does not require the dependent variable to be
and more than 60 years old (0.9%). Nonetheless, the results and findings
continuous and normally distributed. Hence, Mann-Whitney U test was
obtained from the collected sample would provide useful insight into the
carried out on the independent groups when the dependent variables
effects of the pandemic on the travel pattern, particularly for younger
were ordinal, e.g., effect of gender on the number of outdoor trips done
and well-educated people in a developing country.
for non-commuting purposes. The correlations between continuous/
ordinal variables e.g., monthly household income and number of non-
commuting trips was determined using Spearman Correlation. 4.2. Travel pattern before and during covid-19 pandemic

3.3.3. Analysis on the factors influencing mode preferences 4.2.1. Primary purpose of traveling
Section 3 of the questionnaire consisted of items affecting mode In this study, primary traveling purpose is defined as the most
preferences. These items represent certain underlying variables. important cause for which people travel during their everyday life.
Exploratory factor analysis (EFA) is a technique to find out those un­ These may be the trips over which people have very little or no control.
derlying variables also called factors. Hence, EFA was conducted on the For instance, people may reduce their trips for other purposes, but they
items influencing mode preferences during the COVID-19 pandemic to might be compelled to make trips for some primary reasons. People may
extract the underlying factor. Once the underlying factor was discov­ have different primary purposes of traveling; some might travel pri­
ered, the factor scores were computed to represent the relative standing marily for work while others may travel primarily for education etc. The
of each respondent. Regression factor scores were computed and used primary purpose of traveling might vary based on the circumstances e.
for further statistical analyses in this study. g., during a pandemic.
About 55% of the respondents declared that their primary traveling
4. Results and discussion purpose was work before the pandemic; whereas a reduced percentage i.
e., 51% declared that their primary of purpose of traveling was work
4.1. Socio-economic demographics of the respondents during the pandemic (Fig. 2). It could be explained by the fact that work
from home gained popularity during the pandemic. Similarly, the per­
Socio-economic demographics of 671 respondents, whose complete centage of respondents declaring education as primary purpose also
responses were received during a time span of about three weeks, are reduced from 30% before the pandemic to 22% during the pandemic,
shown in Table 1. Most respondents were male (73.3%), with age which was expected since many educational institutes either switched
bracket of 18–30 years (65.9%) and education level of master’s and completely or partially to online education. About 10% of the re­
above (52.2%). Some groups such as male, young and those with higher spondents declared shopping as their primary traveling purpose during
education have a higher share in the collected sample and may produce the pandemic, whereas only about 4% of the respondents declared it as
bias in the results and derived findings. However, it should be noted that their primary purpose before the pandemic.

24
M. Abdullah et al. Transport Policy 108 (2021) 21–33

“Visiting relatives” and “others” were put together into a category

named “others” for further analyses. The results of the McNemar-
Bowker test indicated significant variation between primary traveling
purpose before and during COVID-19 pandemic (χ2 (6) = 40.389, p =
0.000). The cross-tabulation table and McNemar post-hoc test results are
shown in Fig. 3. α and n for Bonferroni correction were set as 0.05 and 6,
respectively. The results indicated that the primary traveling purpose
significantly shifted from work and study trips to shopping trips during
the pandemic. However, expected count in 1 cell (25%) was less than 5
in both cases, therefore, the results should be interpreted with caution.
Also, there was a significant change from study trips to other trips. The
reduction in work and study trips is associated with the work-from-home
and online education policies (de Haas et al., 2020).
Although study trips reduced during the pandemic, they still repre­
sent a considerable portion of the total trips (i.e., 22%). It could be
attributed to the fact that this survey was conducted soon after the
lockdown was over implying that educational institutions had gradually
started their operations. Even during the lockdown, students living in
certain areas did not have access to the internet. Therefore, they had to
travel to nearby places to get internet access in order to join the online
Fig. 2. Distribution of primary traveling purpose before and during COVID-
classes. These trips may also represent a considerable portion of the
19 pandemic.
study trips during the pandemic.

Fig. 3. Change in primary traveling purpose before and during COVID-19.

25
M. Abdullah et al. Transport Policy 108 (2021) 21–33

4.2.2. Frequency of non-commuting trips had no significant effect on number of trips made.
Non-commuting trips, unlike commuting trips e.g., work or educa­ A very weak but significant correlation was found between number
tion trips, are those trips which are not done on daily basis. These may of non-commuting trips and monthly household income before the
include shopping, social, recreational, and other trips. Fig. 4 shows that pandemic (rs = 0.102, p = 0.008). An even weaker but significant cor­
traveling frequency for non-commuting trips reduced during the COVID- relation was found between number of non-commuting trips and
19 pandemic. About 31% of the respondents did not travel at all during monthly household income during the pandemic (rs = 0.078, p = 0.043).
the pandemic. The results of the Wilcoxon signed ranks test pointed out
statistically significant differences in number of non-commuting trips 4.2.3. Preferred mode for shorter distances (<5 km)
carried out before and during the pandemic (Z = − 14.037, p = 0.000). The distribution of mode choices for shorter distances is shown in
Similar findings have been reported in other parts of the world (Landry Fig. 5. Public transport, office/campus transport and taxi/rickshaw were
et al., 2021). put together into a category entitled “Public Transport” for further an­
Mann-Whitney U test was applied to assess the impacts of SEDs on alyses because they had to follow similar instructions from the relevant
the number of non-commuting trips. Certain categories of socio- authorities during the pandemic. Similarly, walking and bicycling were
demographic variables were combined together due to low percentage put together into a new class entitled “Non-motorized”. McNemar-
of respondents in those categories. For example, all the responses with Bowker test indicated that mode choices were significantly different
age above 30 years were combined into a single category entitled “above for short distances (<5 km) before and during the COVID-19 pandemic
30 years”; undergraduate and below education levels were combined (χ2 (6) = 14.055, p < 0.05). Post-hoc McNemar tests were conducted to
into a category entitled “Undergrad and below”; and businessman, determine how modal shift took place for short distances. The cross-
government employee and private employee were combined together to tabulation table and McNemar test results are shown in Fig. 6.
form a new category entitled “Employed”. The results of Mann Whitney The shift from public transport to private car was non-significant.
U test are shown in Table 2. The reason could be the fact that people might think that a close con­
Males undertook significantly more non-commuting trips than fe­ tact with other people for few minutes might not be that dangerous.
males before and during the pandemic. A reason could be that the Moreover, public transport use is not quite common among car owners
financial responsibility of a household is generally considered to lie on in Pakistan. Therefore, no significant shift was observed due to the
men in Pakistan. In addition, women are often discouraged from trav­ pandemic. The respondents significantly shifted from motorbike to non-
eling outside under the pretext of defending them from sexual abuse motorized modes of travel which is a good sign. It could be because trips
(Rizvi et al., 2014). People in the above 30 years old age group made for shopping significantly increased which are likely to be made to the
significantly more non-commuting trips before and during the pandemic nearby shops. Hagberg and Holmberg (2017) also found that shoppers
as compared to the younger age group. It could be because those above tend to use non-motorized mode of travel for short distance shopping
30 years of age are likely to be married and have more social trips. A study conducted in Shenyang, China also reported that shorter
responsibilities. travel distances are important for low-carbon modes (e.g., walking and
Employed people undertook significantly more non-commuting trips bicycling) (Li et al., 2018).
relative to students during the pandemic, however, this difference was
non-significant before the pandemic. A possible reason for this is the fact 4.2.4. Preferred mode for longer distances (>5 km)
that age is likely to be correlated with employment status as well as The distribution of mode choices for longer distances is shown in
social responsibilities. Those in the above 30 years old age are likely to Fig. 7. Public transport, office/campus transport and taxi/rickshaw were
be employed, married, and have more social responsibilities. In addi­ put together into a category entitled “Public Transport” for further an­
tion, the likelihood of having children increases with age which, in turn, alyses. As expected, very few respondents chose bicycle or walking mode
have been found to promote car use (Ye et al., 2018). for distances longer than 5 km. Therefore, walking and bicycling were
Car owners made significantly more non-commuting trips before and omitted from the statistical analysis. McNemar-Bowker test indicated no
during the pandemic than those who did not own a car. Existing evi­ significant difference between mode choice for longer distances (>5 km)
dence suggests that car ownership is associated with number of outdoor before and during the COVID-19 pandemic (χ 2 (3) = 7.743, p > 0.05).
trips (Sillaparcharn, 2007). However, motorbike and bicycle ownership However, post-hoc McNemar test indicated that a significant modal shift
took place from public transport to private car (Fig. 8) because post-hoc
tests are more focused and have more power to detect significant
differences.
The shift from public transport to private car was significant, which
could be attributed to the fact that inter-provincial and inter-city
transport was mostly banned during the pandemic. Secondly, people
may not feel safe in public transport due to close contact with other
people for several minutes or hours as opposed to travel for shorter
distances where the contact is only for few minutes. Existing evidence
suggests that people are likely to prefer their health over economy. For
example, a panel study conducted in Istanbul, Turkey reported that car
owners preferred to travel by their cars owing to the fear of infection
during the pandemic (Shakibaei et al., 2021). The total number of trips
may have decreased during the pandemic; however, the proportion of
private car use increased especially for longer distances.
The modal shift from public transport to motorbike was non-
significant. It could be explained by the fact that riding a motorbike
for longer distances is not convenient, especially during summers when
this survey was conducted.

Fig. 4. Frequency of non-commuting trips before and during COVID-

19 pandemic.

26
M. Abdullah et al. Transport Policy 108 (2021) 21–33

Table 2
Effects of SEDs on the frequency of non-commuting trips (results of Mann Whitney U Test).
Item Group Before COVID-19 During COVID-19

Mean Ranks U P Mean Ranks U p

Gender Male 354.95 34710.5 0.000 359.60 32421 0.000

Female 283.91 271.12
Age 18–30 321.65 44268 0.003 319.19 43178 0.001
Above 30 363.69 368.45
Education Undergrad & below 146.63 2722.5 0.418 137.75 2545 0.223
Masters & above 161.96 162.54
Profession Student 302.99 44266.5 0.067 296.9 42755.5 0.015
Employed 327.62 331.52
Car ownership Yes 367.27 46147 0.000 371.27 44994 0.000
No 312.49 309.48
Motorbike ownership Yes 345.85 52872 0.136 347.53 52296 0.098
No 325.7 323.94
Bicycle ownership Yes 353.84 23620.5 0.312 360.58 23041 0.186
No 333.38 332.39

higher priority on safety and security, social distance, cleanliness,

infection concern, hand sanitizers’ availability, waiting for less con­
gested vehicle and paying for less congested vehicle during the
pandemic. Interestingly, personal social status still plays a role in mode
choice during the pandemic. It is to be noted that there was no signifi­
cant difference between priority placed on comfort before and during
the pandemic. Hence, certain factors, which are important for mode
choice during normal everyday life, may not be as important during a
pandemic. On the other hand, certain pandemic-related factors for mode
choice might be non-existent during normal everyday life, however,
they may become the most important factors during a pandemic.

4.3.1. Exploratory factor analysis (EFA) on the items influencing mode

preferences during the COVID-19 pandemic
EFA (with principal axis factoring) on the items influencing mode
preferences during the COVID-19 pandemic produced a single factor
based on the eigenvalues criteria (eigenvalues > 1). This underlying
factor accounted for about 61.962% of the total variance. The factor
loadings are presented in Table 5 (factor loading cut-off value = 0.40).
The sampling adequacy was satisfactory (Kaiser-Meyer-Olkin measure
= 0.991); the determinant of the matrix was 0.001; and Bartlett’s test of
Fig. 5. Mode choices for shorter distances before and during COVID- sphericity was significant (0.000). Cronbach’s alpha was also satisfac­
19 pandemic. tory (i.e., >0.7).
The regression approach was used to calculate the factor scores,
4.3. Factors influencing mode preferences before and during the COVID- which were subsequently used for determining the impacts of SEDs on
19 pandemic the underlying factor influencing mode preferences during the
pandemic.
There are several factors that influence mode choice under normal
circumstances e.g., travel time, cost, and comfort etc. However, addi­ 4.3.2. Effects of SEDs on the underlying factor influencing mode preferences
tional factors may come into play during a pandemic such as infection during the COVID-19 pandemic
concern, social distance, and hand sanitizer availability in the bus etc. Mann-Whitney U test was carried out to determine the effects of SEDs
The distribution of responses for several factors influencing mode pref­ on the underlying factor influencing mode choice during the pandemic
erences before as well as during the pandemic are shown in Figs. 9 and (Table 6). Females placed more priority on the pandemic-related items
10, respectively. relative to the males during the pandemic. Females have been found to
As expected, most of the respondents put more priority on the perceive more risk of the pandemic as compared to males (Dryhurst
pandemic-related items while choosing a mode during the pandemic. et al., 2020; Rana et al., 2021). Those who did not own a motorbike or a
The mean and standard deviation of the priorities for each item are bicycle put significantly more priority on the items associated with the
shown in Table 3. Mean of the priority placed on each item showed that pandemic. A probable explanation is that such users are likely to use
the respondents put the most priority on cleanliness followed by infec­ public transport, therefore, they might be more worried about the
tion concern during the pandemic. However, they placed the most pri­ pandemic-related items. A weak significant correlation was found be­
ority on comfort followed by safety and security before the pandemic. tween the underlying factor and monthly household income (rs = 0.078,
The least mean priority was placed on comfort during the pandemic and p = 0.046).
hand sanitizer’s availability in public transport vehicles before the
pandemic. 5. Policy implications
In order to determine how people placed priorities on different fac­
tors before and during the pandemic, Wilcoxon Signed Rank tests were 5.1. Shopping as primary purpose of traveling
conducted. The results, as shown in Table 4, indicated that people put
About 10% respondents declared that their primary traveling

27
M. Abdullah et al. Transport Policy 108 (2021) 21–33

Fig. 6. Mode shift for short distances due to COVID-19 (results of McNemar test).

and Ghosh (1999) reported that larger families tend to undertake more
grocery trips because they have more needs. This is worrying in the
perspective of the COVID-19 pandemic because close relatives in
Pakistan generally live together in a joint family system implying that
they are likely to make more trips and are at a greater risk of spreading
the virus to more people (Tian et al., 2020). Moreover, Bawa and Ghosh
(1999) reported that people older than 55 make more grocery trips,
which is consistent with the finding of this study that people above 30
years of age make more non-commuting trips. Whereas current data on
fatality rates suggests that older people are more likely to be affected by
the virus (Liu et al., 2020). In addition, consumers may combine their
trips for several purposes in order to reduce their shopping trip fre­
quency during the pandemic as they do in their normal daily lives to
optimize their time spent on shopping (Leszczyc et al., 2004). Hence,
consumers may be attracted by supermarkets with large variety of items
in order to reduce their trip frequency during the pandemic. Although,
supermarkets and stores around the world have taken measures such as
limiting the number of people in the stores at a time, maintaining social
distance between the consumers and other necessary precautions,
panic-buying was also observed during the pandemic (Nicola et al.,
2020) which can lead to undesirable results. Moreover, lack of aware­
Fig. 7. Mode choices for longer distances before and during COVID- ness among Pakistani consumers especially due to the low literacy rate
19 pandemic. may contribute to the transmission of the virus.
A sudden increase in online shopping has been observed due to the
purpose was shopping during the pandemic, whereas only 4% stated pandemic (Ozili and Arun, 2020) and it has the potential to reduce the
that their primary purpose was shopping before the pandemic. Further, outdoor shopping trips. Convenience and accessibility offered by the
statistical analysis indicated a significant shift in primary traveling online shopping have been found to be the distinct characteristics which
purpose from work and study to shopping. Therefore, shopping trips attract consumers to shop electronically (Wolfinbarger and Gilly, 2001;
require special attention in the post COVID-19 world. For instance, Bawa Jarvenpaa and Todd, 1997; Burke, 1997). Lack of mobility such as an

28
M. Abdullah et al. Transport Policy 108 (2021) 21–33

Fig. 8. Mode shift for long distances due to COVID-19 (results of McNemar test).

Table 3
Mean and standard deviation of user rating for each item affecting mode choice.
Items Before COVID-19 During COVID-19

Mean Std. Mean Std.

Deviation Deviation

Safety and security 3.65 1.222 4.20 1.114

Comfort 3.81 1.078 3.80 1.199
Cleanliness 3.61 1.149 4.28 1.044
Infection concern 3.10 1.286 4.22 1.079
Personal social status 2.96 1.196 3.63 1.333
Social distance 2.29 1.274 4.07 1.175
Hand sanitizer in vehicle 2.10 1.313 3.97 1.256
Waiting for less congested 2.65 1.341 3.84 1.307
vehicle
Paying for less congested 2.80 1.381 3.90 1.253
vehicle

Fig. 9. Distribution of responses for the factors influencing mode preferences

before the pandemic.
Table 4
Factors influencing mode preferences before and during the COVID-19
pandemic (Wilcoxon Signed Rank test results).
Item Group Z Asymp. Sig. (2-
tailed)

Safety and security Before and After − 10.469 .000

Comfort COVID-19 -.023 .982
Cleanliness − 13.171 .000
Infection concern − 15.990 .000
Personal social status − 10.979 .000
Social distance − 18.049 .000
Hand sanitizer in vehicle − 18.281 .000
Waiting for less congested − 15.214 .000
vehicle
Paying for less congested − 14.460 .000
vehicle

shopping especially in the developing countries such as social and

technical factors, lack of trust and affection, overpricing, lack of variety,
Fig. 10. Distribution of responses for the factors influencing mode preferences
security risks, and limited internet access etc. (Alam et al., 2016; Bigdeli
during the pandemic.
et al., 2009; Abbad et al., 2011; Mateen et al., 2018; Nazir et al., 2012;
Lawrence and Tar, 2010). Lack of awareness among people is also an
illness, absence of transport or other factors can also give rise to online obstacle to the adoption of online shopping. An initiative about
shopping (y Monsuwé et al., 2004). Hence, consumers have more rea­ providing latest information to the farmers through smart phones was
sons during the pandemic to conveniently shop online due to mobility undertaken by the government of Pakistan. However, majority of the
constraints such as self-isolation, social distancing, and unavailability of farmers did not even know how to use the smart phones for accessing the
transport etc. However, there are certain factors that hinder online

29
M. Abdullah et al. Transport Policy 108 (2021) 21–33

Table 5 bicycle use was significantly higher during COVID-19 pandemic (p <
Exploratory factor analysis of the items influencing mode 0.001). However, the overall share of active modes as well as bicycle
preferences. ownership (12.8%) is still low. The analysis on the factors affecting
Items Factor 1 mode choice showed that people prefer modes with low infection
Social distance .882
chances and more social distance etc. Walking and cycling possess these
Cleanliness .878 attributes. In addition, there are less vehicles on the roads due to the
Infection concern .855 pandemic (Parr, 2020) providing a good opportunity to shift to
Safety & security .820 non-motorized modes of travel. Less motorized traffic during the
Hand sanitizers in vehicles .795
pandemic may also develop a sense of safety among the bicyclists and
Paying for less congested vehicle .783
Waiting for less congested vehicle .780 pedestrians. Furthermore, there was a significant shift from work and
Comfort .627 study trips to shopping trips. Use of private car for shopping can be
Personal social status .613 further reduced and active modes can be encouraged if distances be­
% of variance explained 61.962 tween homes and stores are reduced (Jiao, 2011). This study also found
that people still pay attention to their personal social status while
Cronbach’s alpha 0.931
choosing a mode which indicates less use of non-motorized modes,
especially bicycles. In order to change the perception about
non-motorized transport and promote its use, a consistent effort at
Table 6
community or government level is needed (Buehler et al., 2016).
Effect of SEDs on the underlying factor influencing mode preferences during the
As the world progresses towards promoting the active modes of
COVID-19 pandemic.
travel (i.e., walking and cycling), the planners in Pakistan focus on
Item Group During COVID-19
ensuring smooth traffic flow for private cars (Haque, 2015). Construc­
Mean Ranks U p tion of flyovers and wide roads have discouraged pedestrians and bi­
Gender Male 307.97 32558 0.000 cyclists in big Pakistani cities. A study about walkability in Asian cities
Female 382.51 indicated that 41% of the respondents believe that the facilities for pe­
Age 18–30 328.47 47646 0.929 destrians in their cities are either bad or very bad (Leather, 2011).
Above 30 327.08
Furthermore, 67% of the respondents would shift from walking to
Education Undergrad & below 185.37 2292 0.176
Masters & above 156.24 motorized modes if the walking environments do not improve. Bicycle
Profession Student 302.20 43904 0.350 trips have been declining in Pakistan due to inadequate cycling infra­
Employed 315.92 structure. A survey conducted in Lahore, Pakistan found cultural issues
Car ownership Yes 326.92 52197 0.899 and gender to be the main barriers to biking followed by other hin­
No 328.81
drances related to the environment and infrastructure (Aslam et al.,
Motorbike ownership Yes 303.16 45245 0.001
No 354.16 2018).
Bicycle ownership Yes 280.14 19765.5 0.013 These findings suggest that a developing country like Pakistan,
No 334.94 which mainly focused on car travel in the past, should now focus on non-
motorized modes of travel. The relevant authorities can properly utilize
this opportunity and maintain the share of active modes even after the
relevant information (Khan et al., 2019). Furthermore, English language
pandemic is over. Pucher et al. (2010) presented a review of the policies
of the online shopping websites can also be a barrier. Nationwide pol­
aimed at promoting bicycling and suggested that a comprehensive
icies about teaching English language and relevant courses to young
approach is better than uncoordinated individual measures to promote
students to raise awareness regarding e-commerce and related security
bicycling. Winters et al. (2017) suggest convenient, safe, and
issues can increase awareness about and trust in e-commerce (Abbad
well-connected bicycling and walking infrastructure for promoting
et al., 2011). In addition, the government should aim at increasing
active travel. They also classified the policies in four broad categories:
internet access. Government of Punjab, the most populated province of
society-level policies, city-level policies, route-level policies, and
Pakistan, provided free Wi-Fi hotspot service at several public places but
individual-oriented policies. Considering the current situation under the
the coverage is still limited to few major cities.
pandemic, where there is less motorized traffic and the financial situa­
Online shopping websites and e-vendors should overcome these
tion is not encouraging, it is recommended that individual-oriented
shortcomings and challenges in order to attract the consumers. This
policies be the top priority for short term planning in order to further
could be achieved by maintaining quality, reducing risk factor and de­
increase the share of active modes. Individual-oriented policies are
livery costs, and adding testimonials, ratings, and reviews from the
informational and educational policies targeting individuals or popula­
previous consumers regarding the product quality and services to
tion groups and, may prove to be useful in motivating people to use
convince the consumers and build trust (Abrar et al., 2017; Yousaf et al.,
active modes of travel during the pandemic.
2012). Entry of famous international brands such as Amazon into the
Pakistani e-commerce industry may also increase the competition and
bring about positive changes. Application of other alternatives such as 5.3. Planning for shorter commuting times
OSaaS: Online Shopping as a Service could be explored to solve some of
the problems such as low literacy rate, communication barrier and This study indicates that less people traveled for work and study
limited or no internet access etc. (Khan et al., 2016). An advantage of during the pandemic, however, the percentage is still substantial. People
OSaaS particularly useful during the pandemic is that customers can need to travel for various reasons during a pandemic while maintaining
place their orders to the cloud service center in their local language social distance and adopting other necessary precautions. In addition,
using cell phones. essential workers need to commute even during a lockdown. This study
also found that people shifted from public transport to private car for
longer distances during the pandemic. Major cities of Pakistan have seen
5.2. Non-motorized/active modes for shorter distances ill-planned growth over the years. Moreover, facilities, shopping, and
health etc., are generally separated by long distances (Qasim et al.,
The results of the current study showed that there was a significant 2013). Although these ideas seem workable under lockdown and social
shift from motorbike to walking and bicycling for shorter distances (<5 distancing protocols, such urban planning policies force people to travel
km) during the pandemic. In addition, McNemar test showed that long distances for work, education and other purposes and private car

30
M. Abdullah et al. Transport Policy 108 (2021) 21–33

use is encouraged particularly during a pandemic. 6. Conclusions

The urban planning policies in Pakistan have two major negative
consequences. Firstly, there will be, arguably, higher chances of disease This study explored the effects of the COVID-19 pandemic on travel
spread if public transport is used because active modes are not suitable patterns and mode preferences in a sample from three cities of province
for longer distances. Secondly, people may shift from public transport to Punjab of Pakistan. A questionnaire was administered online to collect
private car for longer distances, as found in this study, resulting in the data about demographics, primary travel behavior, mode choice and
negative environmental impacts. By the late 20th century, the smart factors influencing the mode preferences before and during the
urbanism policies to make cities smaller and compact began to be pandemic. The statistical analyses indicated differences between pri­
implemented (Batty, 2020). Various other policies such as reducing car mary traveling purpose before and during the pandemic. The results
use, transit-oriented-development and road pricing etc. were also indicated that less people traveled for work and study during the
implemented at several places around the world. One of the objectives pandemic as compared to the before-pandemic situation. However,
was to promote healthier and environment friendly modes of transport. more people traveled for shopping during the pandemic. Number of trips
Hence, the compact city idea, aside from being popular owing to its performed for non-commuting purposes were found to be significantly
other advantages during normal life, seems plausible under pandemic different before and during the pandemic. Males undertook significantly
situations. Although pandemics do not happen very often, future urban more non-commuting trips as compared to females during the
planning policies oriented towards the core idea of compact cities may pandemic. Those above 30 years old and car owners made significantly
prove to be beneficial in terms of shorter commuting distances which more non-commuting trips during the pandemic. Monthly household
can be covered by non-motorized modes of travel in shorter times. income was found to be positively correlated with number of non-
commuting trips before and during the pandemic.
Mode choice was significantly different before and during the
5.4. Long term impacts on public transport pandemic for shorter distances (<5 km). Public transport use reduced,
whereas private car use increased during the pandemic. However, the
Since public transport is the worst affected mode of travel during the shift between public transport and private car was not significant for
COVID-19 pandemic, the evolving situation of the pandemic indicates shorter distances. There was a significant modal shift from motorbike to
three possible long term scenarios: 1. people avoid the use of public bicycling and walking. Mode choice was also found to be significantly
transport altogether and it may never recover, 2. people prefer active different for longer distances (>5 km) before and during the pandemic
modes and private car over public transport in the long term. 3. people and the shift from public transport to private car was also statistically
start using public transport as soon as fear of the pandemic and the significant.
pandemic itself go away. The respondents also placed priorities on different factors influ­
Certain initial studies favored the first scenario indicating that public encing the mode preferences before as well as during the COVID-19
transport may not recover any time soon. For example, at the end of pandemic. Mean of the user rating for each factor indicated that the
June 2020, the use of general traffic and shared bikes reached around respondents placed the most priority on cleanliness followed by infec­
85%, while public transport had risen to only about 60% of the use in the tion concern during the pandemic. However, they placed the most pri­
previous reference period in Spain (Orro et al., 2020). However, the ority on comfort followed by safety and security before the pandemic.
experience from the past pandemics and the current prevailing situation Least priority was placed on comfort during the pandemic and hand
indicates that the first scenario seems improbable. sanitizer availability in vehicle before the pandemic. EFA on the factors
The results of this study and the studies conducted, when pandemic influencing mode preferences during the pandemic yielded an under­
had spread widely, supported the second scenario indicating that people lying factor. Statistical analysis indicated that females put higher pri­
may prefer active modes and private transport over public transport. For ority on the pandemic-related items relative to the males during the
example, Conway et al. (2020) reported that many respondents are pandemic. Those who did not own a motorbike, or a bicycle put
planning to walk and ride bicycle more than before (Conway et al., significantly more priority on the pandemic-related items. Monthly
2020). Furthermore, walking and private car use increased and public household income was found to have a very weak positive correlation
transport use declined in two cities in Italy (Moslem et al., 2020). with the underlying factor.
However, multiple studies have recently begun to report the evi­ Based on the results of this study, policy implications regarding
dence for the third scenario. For instance, Przybylowski et al. (2021) shopping as a primary purpose of traveling, mode choice for shorter
reported that 75% of the public transport users in the city of Gdańsk, in distances, shorter commuting times, and long-term impact on public
Poland plan to restart using the public transport once the pandemic is transport are also presented. This study had some limitations. The
over. People are returning to public transport in China, however, no sample was not representative mainly in terms of education level. The
COVID-19 clusters have been found to be associated with public trans­ number of respondents for age groups above 45 years old was very small
port (Shen et al., 2020). Hence, it is recommended that transport au­ probably owing to their less familiarity with online tools and social
thorities focus on improving the perceptions about public transport media. In addition, people may have some discrepancies while reporting
system during the pandemic to attract the users in the long term. their travel behavior few months before the virus emerged. Since the
Furthermore, Beliaev et al. (2020) recommended using financial in­ pandemic is still going on, further studies on this topic may target cross-
centives to recover public transport in the long term while keeping in city and cross-country comparisons to better understand the trans­
view the users’ preferences. Finally, the extent to which public transport ferability of the findings of this study.
recovers also depends on the work-from-home policies, which may This study is based on the findings from a questionnaire study con­
affect the long term travel behavior (Beck and Hensher, 2020). Yet ducted in one country. Yet it is hypothesized that the findings reported
public transport may still remain to be a need of the society because here might be applicable to other societies with similar socio-
there are certain sectors which require morvement from homes to work demographic and economic characteristics, and wherever similar
such as manufacturing, construction, and many service-related jobs. travel patterns and mode choices persist (e.g., the cities in the South
Policies to attract people towards pubic transport are beneficial for the Asian countries i.e., India, Bangladesh, Afghanistan, and Sri Lanka).
society, however, these policies must be developed and implemented There could be some similarities with the cities of the developed coun­
while keeping in view the ethical considerations. For example, public tries, however, caution is advised while transferring these findings to
transport may, arguably, propagate the spread of the virus. Hence, any such cities and further comparative studies are recommended. Since the
policies aimed at reviving public transport must be based on users’ sample is skewed towards male, younger and more educated people, the
preferences. findings should be applied to populations with similar characteristics.

31
M. Abdullah et al. Transport Policy 108 (2021) 21–33

The results presented here might not be applicable to cities where EU, 2020. COVID-19: Guidelines on the Progressive Restoration of Transport Services
and Connectivity. European Union.
transport modes are different. In addition, the scale of the pandemic
Gibbons, J.D., Chakraborti, S., 2014. Nonparametric Statistical Inference: Revised and
should also be considered while generalizing the results of this study. Expanded. CRC press.
Hagberg, J., Holmberg, U., 2017. Travel modes in grocery shopping. Int. J. Retail Distrib.
Manag. 45 (9), 991–1010.
Author statement Hamer, M., Chida, Y., 2008. Walking and primary prevention: a meta-analysis of
prospective cohort studies. Br. J. Sports Med. 42 (4), 238–243.
Muhammad Abdullah: Conceptualization, Methodology, Formal Haque, N.U., 2015. Flawed Urban Development Policies in Pakistan. Pakistan Institute of
Development Economics, Islamabad.
analysis, Investigation, Writing - Original Draft, Review & Editing, Ibuka, Y., Chapman, G.B., Meyers, L.A., Li, M., Galvani, A.P., 2010. The dynamics of risk
Visualization. perceptions and precautionary behavior in response to 2009 (H1N1) pandemic
Nazam Ali: Conceptualization, Methodology, Writing - Original influenza. BMC Infect. Dis. 10 (1), 1–11.
ITE, 2020. COVID-19 Resources. Institute of Transportation Engineers. Available at: https
Draft, Review & Editing. ://www.ite.org/about-ite/covid-19-resources/?_zs=MHVdl&_zl=qbii1. (Accessed
Syed Arif Hussain: Conceptualization, Methodology, Writing - Orig­ 22 June 2020).
inal Draft, Review & Editing. Jarvenpaa, S.L., Todd, P.A., 1997. Is there a future for retailing on the Internet.
Electronic marketing and the consumer 1 (12), 139–154.
Atif Bilal Aslam: Conceptualization, Methodology, Writing - Original
Jiao, J., Moudon, A.V., Drewnowski, A., 2011. Grocery shopping: how individuals and
Draft, Review & Editing. built environments influence choice of travel mode. Transport. Res. Rec. 2230 (1),
Muhammad Ashraf Javid: Conceptualization, Methodology, Writing 85–95.
Kampf, G., Todt, D., Pfaender, S., Steinmann, E., 2020. Persistence of coronaviruses on
- Original Draft, Review & Editing.
inanimate surfaces and their inactivation with biocidal agents. J. Hosp. Infect. 104
(3), 246–251.
Appendix A. Supplementary data Khan, M., Xu, X., Dou, W., Yu, S., 2016. December. OSaaS: online shopping as a service to
escalate E-commerce in developing countries. In: 2016 IEEE 18th International
Conference on High Performance Computing and Communications; IEEE 14th
Supplementary data to this article can be found online at https://doi. International Conference on Smart City; IEEE 2nd International Conference on Data
org/10.1016/j.tranpol.2021.04.023. Science and Systems (HPCC/SmartCity/DSS). IEEE, pp. 1402–1409.
Khan, N.A., Qijie, G., Ali, S., Shahbaz, B., Shah, A.A., 2019. Farmers’ use of mobile phone
for accessing agricultural information in Pakistan. Ciência Rural. 49 (10).
References Kim, C., Cheon, S.H., Choi, K., Joh, C.H., Lee, H.J., 2017. Exposure to fear: changes in
travel behavior during MERS outbreak in Seoul. KSCE Journal of Civil Engineering
21 (7), 2888–2895.
Abbad, M., Abbad, R., Saleh, M., 2011. Limitations of E-commerce in Developing
Kitchen, C.M., 2009. Nonparametric vs parametric tests of location in biomedical
Countries: Jordan Case. Education, Business and Society: Contemporary Middle
research. Am. J. Ophthalmol. 147 (4), 571–572.
Eastern Issues.
Kraemer, M.U., Yang, C.H., Gutierrez, B., Wu, C.H., Klein, B., Pigott, D.M., Du Plessis, L.,
Abrar, K., Naveed, M., Ramay, M.I., 2017. Impact of perceived risk on online impulse
Faria, N.R., Li, R., Hanage, W.P., Brownstein, J.S., 2020. The effect of human
buying tendency: an empirical study in the consumer market of Pakistan. J. Account.
mobility and control measures on the COVID-19 epidemic in China. Science 368
Market. 6 (3), 246.
(6490), 493–497.
Alam, A., Malik, O.M., Hadi, N.U., Gaadar, K., 2016. Barriers of online shopping in
Landry, C.E., Bergstrom, J., Salazar, J., Turner, D., 2021. How has the COVID-19
developing countries: case study of Saudi Arabia. Eur. Acad. Res. 3 (12),
pandemic affected outdoor recreation in the US? A revealed preference approach.
12957–12971.
Appl. Econ. Perspect. Pol. 43 (1), 443–457.
APTA, 2020. The COVID-19 Pandemic Public Transportation Responds: Safeguarding
Lawrence, J.E., Tar, U.A., 2010. Barriers to e-commerce in developing countries. Inf. Soc.
Riders and Employees. American Public Transportation Association.
Jus. J. 3 (1), 23–35.
Aslam, S.A.B., Masoumi, H.E., Asim, M., Anwer, I., 2018. Cyclability in Lahore, Pakistan:
Leather, J., Fabian, H., Gota, S., 2011. Walkability and Pedestrian Facilities in Asian
looking into potential for greener urban traveling. TeMA-Journal of Land Use,
Cities State and Issues.
Mobility and Environment 11 (3), 323–343.
Leszczyc, P.T.P., Sinha, A., Sahgal, A., 2004. The effect of multi-purpose shopping on
Batty, M., 2020. The Coronavirus Crisis: what Will the Post-pandemic City Look like?.
pricing and location strategy for grocery stores. J. Retailing 80 (2), 85–99.
Bawa, K., Ghosh, A., 1999. A model of household grocery shopping behavior. Market.
Li, J., Lo, K., Guo, M., 2018. Do socio-economic characteristics affect travel behavior? A
Lett. 10 (2), 149–160.
comparative study of low-carbon and non-low-carbon shopping travel in Shenyang
Beck, M.J., Hensher, D.A., 2020. Insights into the impact of COVID-19 on household
City, China. Int. J. Environ. Res. Publ. Health 15 (7), 1346.
travel and activities in Australia–The early days under restrictions. Transport Pol. 96,
Litman, T., 2020. Lessons from pandemics: valuing public transportation. Planetizen.
76–93.
Available at: https://www.planetizen.com/blogs/109584-lessons-pandemics-valu
Beliaev, M., Bıyık, E., Lazar, D.A., Wang, W.Z., Sadigh, D., Pedarsani, R., 2020.
ing-public-transportation. (Accessed 20 June 2020).
Incentivizing Routing Choices for Safe and Efficient Transportation in the Face of the
Liu, K., Chen, Y., Lin, R., Han, K., 2020. Clinical features of COVID-19 in elderly patients:
COVID-19 Pandemic. arXiv preprint arXiv:2012.15749. (Unpublished).
a comparison with young and middle-aged patients. J. Infect. 80 (6), e14–e18.
Bigdeli, O., Afaghzadeh, S., Albadvi, A., Abdolvand, N., 2009. Barriers of online shopping
Mateen Khan, M., Ahmed, E., 2018. Online grocery shopping and consumer perception: a
in developing countries: case study Iran. In: Proceedings of the IADIS International
case of Karachi market in Pakistan. In: Saleem, M., Khan, M.M., Ahmed, M.E., Ali, S.,
Conference E-Commerce, pp. 3–10.
Shah, N., Rafiq, S. (Eds.), (2018). Online Grocery Shopping and Consumer
Browne, A., St-Onge Ahmad, S., Beck, C.R., Nguyen-Van-Tam, J.S., 2016. The roles of
Perception: A Case of Karachi Market in Pakistan, pp. 1–13. Journal of Internet and
transportation and transportation hubs in the propagation of influenza and
e-Business Studies.
coronaviruses: a systematic review. J. Trav. Med. 23 (1), tav002.
McCloskey, B., Heymann, D.L., 2020. SARS to novel coronavirus–old lessons and new
Buehler, R., Götschi, T., Winters, M., 2016. Moving toward Active Transportation: How
lessons. Epidemiol. Infect. 148.
Policies Can Encourage Walking and Bicycling.
Milne, G.J., Kelso, J.K., Kelly, H.A., Huband, S.T., McVernon, J., 2008. A small
Burke, R.R., 1997. Do you see what I see? The future of virtual shopping. J. Acad.
community model for the transmission of infectious diseases: comparison of school
Market. Sci. 25 (4), 352–360.
closure as an intervention in individual-based models of an influenza pandemic. PloS
Cahyanto, I., Wiblishauser, M., Pennington-Gray, L., Schroeder, A., 2016. The dynamics
One 3 (12), e4005.
of travel avoidance: the case of Ebola in the US. Tourism Management Perspectives
Mogaji, E., 2020. Impact of COVID-19 on transportation in lagos, Nigeria. Transportation
20, 195–203.
research interdisciplinary perspectives 6, 100154.
Conway, M.W., Salon, D., da Silva, D.C., Mirtich, L., 2020. How will the COVID-19
Moslem, S., Campisi, T., Szmelter-Jarosz, A., Duleba, S., Nahiduzzaman, K.M.,
pandemic affect the future of urban life? Early evidence from highly-educated
Tesoriere, G., 2020. Best–worst method for modelling mobility choice after COVID-
respondents in the United States. Urban Science 4 (4), 50.
19: evidence from Italy. Sustainability 12 (17), 6824.
Cooley, P., Brown, S., Cajka, J., Chasteen, B., Ganapathi, L., Grefenstette, J.,
Musselwhite, C., Avineri, E., Susilo, Y., 2020. Editorial JTH 16–The Coronavirus Disease
Hollingsworth, C.R., Lee, B.Y., Levine, B., Wheaton, W.D., Wagener, D.K., 2011. The
COVID-19 and implications for transport and health. Journal of Transport & Health
role of subway travel in an influenza epidemic: a New York City simulation. J. Urban
16, 100853.
Health 88 (5), 982.
Nazir, S., Tayyab, A., Sajid, A., ur Rashid, H., Javed, I., 2012. How online shopping is
de Haas, M., Faber, R., Hamersma, M., 2020. How COVID-19 and the Dutch ‘intelligent
affecting consumers buying behavior in Pakistan? International Journal of Computer
lockdown’change activities, work and travel behaviour: evidence from longitudinal
Science Issues (IJCSI) 9 (3), 486.
data in The Netherlands. Transportation Research Interdisciplinary Perspectives 6,
Nicola, M., Alsafi, Z., Sohrabi, C., Kerwan, A., Al-Jabir, A., Iosifidis, C., Agha, M.,
100150.
Agha, R., 2020. The socio-economic implications of the coronavirus and COVID-19
Dryhurst, S., Schneider, C.R., Kerr, J., Freeman, A.L., Recchia, G., Van Der Bles, A.M.,
pandemic: a review. Int. J. Surg. 78, 185–193.
Spiegelhalter, D., van der Linden, S., 2020. Risk perceptions of COVID-19 around the
Orro, A., Novales, M., Monteagudo, Á., Pérez-López, J.B., Bugarín, M.R., 2020. Impact on
world. J. Risk Res. 23 (7–8), 994–1006.
city bus transit services of the COVID–19 lockdown and return to the new Normal:
Edelson, P.J., Phypers, M., 2011. TB transmission on public transportation: a review of
the case of A Coruña (Spain). Sustainability 12 (17), 7206.
published studies and recommendations for contact tracing. Trav. Med. Infect. Dis. 9
(1), 27–31.

32
M. Abdullah et al. Transport Policy 108 (2021) 21–33

Otsuki, S., Nishiura, H., 2016. Reduced risk of importing Ebola virus disease because of Transit Center, 2020. How transit agencies are responding to the COVID-19 public health
travel restrictions in 2014: a retrospective epidemiological modeling study. PloS One threat. Transit Center. Available at: https://transitcenter.org/how-transit-agencies-
11 (9), e0163418. are-responding-to-the-covid-19-public-health-threat/. (Accessed 22 June 2020).
OurWorldinData, 2020. Pakistan: coronavirus pandemic. https://ourworldindata. Troko, J., Myles, P., Gibson, J., Hashim, A., Enstone, J., Kingdon, S., Packham, C.,
org/coronavirus/country/pakistan?country=~PAK. (Accessed 17 November 2020). Amin, S., Hayward, A., Van-Tam, J.N., 2011. Is public transport a risk factor for
Ozili, P.K., Arun, T., 2020. Spillover of COVID-19: Impact on the Global Economy. acute respiratory infection? BMC Infect. Dis. 11 (1), 1–6.
Available at SSRN 3562570. TUMI, 2020. The COVID-19 outbreak and implications to sustainable urban mobility –
O’Sullivan, F., 2020. In Japan and France, riding transit looks surprisingly safe. some observations.Transformative Urban Mobility Initiative. Available at: https
Bloomberg CityLab. Available at: https://www.bloomberg.com/news/articles/2020 ://www.transformative-mobility.org/news/the-covid-19-outbreak-and-implications
-06-09/japan-and-france-find-public-transit-seems-safe. (Accessed 22 June 2020). -to-public-transport-some-observations. (Accessed 22 June 2020).
Parr, S., Wolshon, B., Renne, J., Murray-Tuite, P., Kim, K., 2020. Traffic impacts of the UITP, 2020. Cities for People Means Prioritising Public Transport. International
COVID-19 pandemic: statewide analysis of social separation and activity restriction. Association of Public Transport. Available at:https://www.uitp.org/news/cities-p
Nat. Hazards Rev. 21 (3), 04020025. eople-means-prioritising-public-transport. (Accessed 22 June 2020).
PBIC, 2020. Local actions to support walking and cycling during social distancing Van Doremalen, N., Bushmaker, T., Morris, D.H., Holbrook, M.G., Gamble, A.,
dataset.Pedestrian and Bicycle Information Center. Available at: http://pedbikeinfo. Williamson, B.N., Tamin, A., Harcourt, J.L., Thornburg, N.J., Gerber, S.I., Lloyd-
org/resources/resources_details.cfm?id=5209. (Accessed 22 June 2020). Smith, J.O., 2020. Aerosol and surface stability of SARS-CoV-2 as compared with
Peeri, N.C., Shrestha, N., Rahman, M.S., Zaki, R., Tan, Z., Bibi, S., Baghbanzadeh, M., SARS-CoV-1. N. Engl. J. Med. 382 (16), 1564–1567.
Aghamohammadi, N., Zhang, W., Haque, U., 2020. The SARS, MERS and novel West, R., Michie, S., Rubin, G.J., Amlôt, R., 2020. Applying principles of behaviour
coronavirus (COVID-19) epidemics, the newest and biggest global health threats: change to reduce SARS-CoV-2 transmission. Nature Human Behaviour 4 (5),
what lessons have we learned? Int. J. Epidemiol. 49 (3), 717–726. 451–459.
Polis, 2020. COVID-19: keeping things moving. Cities and regions for transport WHO, 2020. Coronavirus disease (COVID-19) pandemic. Retrieved 74, 2020, from. http
innovation. Available at: https://www.polisnetwork.eu/document/covid-19-kee s://www.who.int/emergencies/diseases/novel-coronavirus-2019.
ping-things-moving/. (Accessed 22 June 2020). Wilder-Smith, A., Chiew, C.J., Lee, V.J., 2020. Can we contain the COVID-19 outbreak
Przybylowski, A., Stelmak, S., Suchanek, M., 2021. Mobility behaviour in view of the with the same measures as for SARS? Lancet Infect. Dis. 20 (5), e102–e107.
impact of the COVID-19 pandemic—public transport users in gdansk case study. Williams, C.J., Schweiger, B., Diner, G., Gerlach, F., Haaman, F., Krause, G.,
Sustainability 13 (1), 364. Nienhaus, A., Buchholz, U., 2010. Seasonal influenza risk in hospital healthcare
Pucher, J., Dill, J., Handy, S., 2010. Infrastructure, programs, and policies to increase workers is more strongly associated with household than occupational exposures:
bicycling: an international review. Prev. Med. 50, S106–S125. results from a prospective cohort study in Berlin, Germany, 2006/07. BMC Infect.
Qasim, Mohammad, Zaidi, S.S., 2013. Ensuring sustainable development through urban Dis. 10 (1), 1–11.
planning in Pakistan. Mehran Univ. Res. J. Eng. Technol. 32 (2), 207–220. Winters, M., Buehler, R., Götschi, T., 2017. Policies to promote active travel: evidence
Rana, I.A., Bhatti, S.S., Aslam, A.B., Jamshed, A., Ahmad, J., Shah, A.A., 2021. COVID-19 from reviews of the literature. Curr. Environ. Health Rep. 4 (3), 278–285.
risk perception and coping mechanisms: does gender make a difference? Wolfinbarger, M., Gilly, M.C., 2001. Shopping online for freedom, control, and fun. Calif.
International Journal of Disaster Risk Reduction 55, 102096. Manag. Rev. 43 (2), 34–55.
Reinhardt-Rutland, A., 2011. Walking and Cycling: Local Measures to Promote Walking Woodcock, J., Wright, J., Whitelegg, J., Watson, P., Walters, H., Walker, I., Uttley, J.,
and Cycling as Forms of Traffic or Recreation (PH41). Tulley, I., 2020. Researchers call on government to enable safe walking and cycling
Rizvi, N., Khan, K.S., Shaikh, B.T., 2014. Gender: shaping personality, lives and health of during the COVID-19 pandemic. Available at: https://docs.google.com/document/
women in Pakistan. BMC Wom. Health 14 (1), 1–8. d/e/2PACX-1vR5AdOmF2effrg-lpBXtvh0stbxM0W6xTDwV2J-xIgHB8rPfZl5bLV
Shakibaei, S., De Jong, G.C., Alpkökin, P., Rashidi, T.H., 2021. Impact of the COVID-19 R5eL7VV2m_W9xx5PgH26TB0vq/pub. (Accessed 20 March 2020).
pandemic on travel behavior in Istanbul: a panel data analysis. Sustainable cities and Wu, A., Peng, Y., Huang, B., Ding, X., Wang, X., Niu, P., Meng, J., Zhu, Z., Zhang, Z.,
society 65, 102619. Wang, J., Sheng, J., 2020. Genome composition and divergence of the novel
Shen, J., Duan, H., Zhang, B., Wang, J., Ji, J.S., Wang, J., Pan, L., Wang, X., Zhao, K., coronavirus (2019-nCoV) originating in China. Cell Host Microbe 27 (3), 325–328.
Ying, B., Tang, S., 2020. Prevention and Control of COVID-19 in Public y Monsuwé, T.P., Dellaert, B.G., De Ruyter, K., 2004. What drives consumers to shop
Transportation: Experience from China. Environmental pollution, p. 115291. online? A literature review. Int. J. Serv. Ind. Manag. 15 (1), 102–121.
Sillaparcharn, P., 2007. Vehicle ownership and trip generation modelling: a case study of Ye, N., Gao, L., Juan, Z., Ni, A., 2018. Are people from households with children more
Thailand. IATSS Res. 31 (2), 17–26. likely to travel by car? An empirical investigation of individual travel mode choices
SLOCAT, 2020. COVID-19 and the sustainable transport Community.Sustainable low in Shanghai, China. Sustainability 10 (12), 4573.
carbon transport partnership. Available at: https://slocat.net/covid-19-and-th Yousaf, U., Altaf, M., Sarwar, N., Hassan Shah, S.A., 2012. Hesitancy towards online
e-sustainable-transport-community/#overall. (Accessed 22 June 2020). shopping, A study of Pakistani consumers. Management & Marketing Journal 10 (2).
Stokell, T., 2020. Beyond the lockdown – planning for transportation in the Coronavirus Zhang, Y., Zhang, Y., Liu, Z., 2011. June. The role of different transportation in the
era. Ride Healthy. Available at: https://www.ride-healthy.com/beyond-the-lockdow spreading of new pandemic influenza in mainland China. In: 2011 19th International
n. (Accessed 22 June 2020). Conference on Geoinformatics. IEEE, pp. 1–6.
Tian, S., Hu, N., Lou, J., Chen, K., Kang, X., Xiang, Z., Chen, H., Wang, D., Liu, N., Liu, D., Zhang, Y., Zhang, A., Wang, J., 2020. Exploring the roles of high-speed train, air and
Chen, G., 2020. Characteristics of COVID-19 infection in beijing. J. Infect. 80 (4), coach services in the spread of COVID-19 in China. Transport Pol. 94, 34–42.
401–406.

33