Asian Transport Studies 9 (2023) 100099

Contents lists available at ScienceDirect

Asian Transport Studies

journal homepage: www.sciencedirect.com/journal/asian-transport-studies

Impact of COVID-19 on public transport usage in an anticipated ‘new

normal’ situation: The case of a South Asian country based on first
wave data
Niaz Mahmud Zafri a, Asif Khan a, Shaila Jamal b, *, Bhuiyan Monwar Alam c
a
Department of Urban and Regional Planning, Bangladesh University of Engineering and Technology, Dhaka, 1000, Bangladesh
b
McMaster University: School of Earth, Environment and Society, McMaster University, 1280 Main St. West, Hamilton, ON, L8S 4K1, Canada
c
Department of Geography & Planning, University of Toledo, Ohio, 43606, USA

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

Keywords: This study attempted to investigate the impacts of COVID-19 on public transport usage in a hypothetical ‘new
COVID-19 normal’ situation in a South Asian country, Bangladesh, by using data collected during the first wave of COVID-
Public transportation 19. Most of the samples came from young and affluent groups. Findings of the study showed that a substantial
New normal
proportion of respondents expected to reduce travel by public transport during the ‘new normal’ situation than
Lockdown
South Asia
the pre-pandemic situation. To identify the factors behind the expectation, a multinomial logistic regression
model was developed. Results suggest that income, regular travel mode, frequency of travel by public transport
in the pre-COVID-19 situation, expected change in trip frequency and virtual activities, risk perception, and trust
in preventive strategies can influence public transport use during the ‘new normal’ situation. The results of the
study would be useful in understanding the immediate impact of a pandemic on public transportation and help
prepare better for future pandemics.

1. Background showed that travelers’ perceive public transportation as the riskier mode
for COVID-19 transmission and active and private modes as a safer one
The COVID-19 pandemic has had an unprecedented effect not only (Zafri et al., 2022; Barbieri et al., 2021). Therefore, it is highly likely that
on the economy but also on different facets of people’s lives. The world the fear for personal safety may influence public transport users to travel
was hardly prepared when the COVID-19 pandemic started. After being less or use other perceived safer modes after the initial lockdown period
first detected in the city of Wuhan in China in late December 2019, the (De Vos, 2020).
novel coronavirus, which causes the COVID-19 disease, has spread to Studies suggest that trip generation has decreased in many devel­
almost all the countries in the world (Jiang et al., 2020; Lipsitch et al., oping countries, including Chile, China, Bangladesh, and India (Astroza
2020). et al., 2020; Abdullah et al., 2020; Ipsos, 2020) during the first wave of
Usually, the use of public transport depends on factors such as fare, the pandemic. Among all the urban transport modes, public transport
quality of service, travelers’ income, age, gender, car ownership, land suffered the most. As people started to work from home and students
use type, built environment, public transport accessibility, safety, and joined virtual classes, the need for mobility regressed, resulting in a
security (e.g., Paulley et al. 2006; van Lierop et al. 2018; van Soest et al. decline in the demand for public transport by 80–90% in major cities of
2020; An et al., 2019). The COVID-19 pandemic has made travelers’ China and Iran (UITP, 2020; Gkiotsalitis and Cats, 2021). Anwari et al.
modal choice behavior a complex phenomenon, especially in terms of (2021) found that the overall demand for public transport declined by
using public transport. For example, personal safety measures to mini­ more than 50% in Bangladesh during the early stage of the pandemic.
mize the chance of being infected by COVID-19 include wearing a face Another study by Abdullah et al. (2020), based on a survey of 1203
mask, sanitizing the hands, and keeping a distance of 1–2 m from others, people spread over 15 countries, including seven developing countries,
which are very difficult to maintain inside a public transport vehicle reported that 36% of the respondents were using public transportation
(Yezli and Khan, 2020; Tirachini and Cats, 2020). Previous studies for their primary trip purpose before the pandemic, which reduced to

* Corresponding author.
E-mail addresses: [email protected] (N.M. Zafri), [email protected] (A. Khan), [email protected] (S. Jamal), [email protected] (B.M. Alam).

https://doi.org/10.1016/j.eastsj.2023.100099
Received 31 March 2022; Received in revised form 19 December 2022; Accepted 15 February 2023
Available online 17 February 2023
2185-5560/© 2023 The Authors. Published by Elsevier Ltd on behalf of Eastern Asia Society for Transportation Studies. This is an open access article under the CC
BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
N.M. Zafri et al. Asian Transport Studies 9 (2023) 100099

13% during the early stage of the pandemic. Studies conducted in such as a subway system, metro rail, or light rail. The public transport
developing countries like Ghana (Sogbe, 2021) and Nigeria (Mogaji, service was also suspended on March 26, 2020. On June 1, 2020, the
2020) also showed that the trip share of public transport declined due to service resumed subject to adopting and following some
COVID-19. government-mandated recommendations and safety measures (Anwari
It should also be noted that the vaccine was not available during the et al., 2021). For example, buses were not allowed to take passengers
first wave of the pandemic. It might have led people to think that they more than 50% of their seating capacity. Bus operators needed to pro­
have to get accustomed to dealing with the virus and carry out their vide hand sanitizers. Everybody on board, including the staff and pas­
activities in the future in a situation termed as ‘new normal.’ This sengers, was required to wear face masks. In lieu of a reduction in fare
perception regarding the effect of COVID-19 could shape the public collection due to restrictions on the number of passengers, operators
decision and attitude regarding long-term travel behavior, especially were allowed to charge 60% excess fare. As the operators were not
mode preference. For instance, individuals may feel public transport following the government recommended guidelines and were taking
unsafe, and therefore, instead of buying a bicycle that is environment more passengers than allowed while charging 60% more fare, the limit
friendly, they might buy a car or motorcycle, which could change the on passenger numbers was increased to full-seating capacity from
travel behavior of the people in the long run (Luan et al., 2021). A study September 1, 2020. The fare was brought down to the rate that was in
by Zafri et al. (2021b) showed that about 46% of the respondents were place in the pre-COVID-19 situation. This arrangement continued until
expected to increase travel by motorcycle in the post-lockdown period the second wave of COVID-19 hit mid-March 2021.
compared to the pre-COVID-19 situation in Dhaka, Bangladesh and 31%
of the respondents were planning to purchase a motorcycle within 2.2. Data collection and descriptive statistics
August 2021. Unfortunately, it would be a backward shift in achieving
sustainable transportation outcomes if individuals start to make a modal A web-based survey was conducted from July to August 2020 as in-
shift from public transport to privately owned vehicles. Therefore, to person/face-to-face surveys would have violated the pandemic-related
ensure sustainability in the transportation sector, new policies and physical distancing restrictions. Anyone interested was welcomed to
strategies need to be developed for public transport services to prevent participate in the study. A total of 804 samples were collected through
this potential switch of transport modes. So, it would be worthwhile to the survey from all over Bangladesh.
investigate how travelers during the initial period of the pandemic After constructing the questionnaire, five surveyors were hired to
perceived traveling through public transportation in the ‘new normal’ disseminate the survey through various electronic means such as the
situation. internet, social media sites, and emails. Also, a widespread social media
Several studies have looked into the impact of COVID-19 on travel campaign (e.g., Facebook advertisement) was conducted to reach a
behavior, including mode choice, in the context of developing countries broader audience throughout Bangladesh. The survey was promoted on
during the early stage of the pandemic (Anwari et al., 2021; Abdullah various social media sites and groups by utilizing personal and profes­
et al., 2020; Zafri et al., 2021; Bhaduri et al., 2020; Paul et al. 2022a, sional networks as well as by directly contacting the administrators of
2022b). A few studies have also tried to anticipate the future travel social media groups. Both English and Bengali versions of the survey
scenario during post-pandemic situation (long term effect of COVID-19 were available to the respondents.
on travel behavior) (e.g., Hensher et al., 2021; Zhang and Zhang, 2021; The survey collected information on respondents’ socio-
Mogaji et al., 2022; Gkiotsalitis and Cats, 2021). However, travelers’ demographic characteristics such as age, gender, income, and living
anticipation at the early stage of the pandemic on expected public arrangements. A range of travel behavior-related questions such as
transport usage during the ‘new normal’ situation was rarely examined regular transport mode, frequency of using public transport, and vehicle
in the existing literature, especially in the context of a developing ownership (e.g., car, motorcycle, bicycle) was included in the ques­
country. This study explored this gap in the literature in the context of tionnaire. A set of perception-based questions were asked to understand
developing countries by using Bangladesh, a South Asian country, as the how the respondents perceived the COVID–19 transmission risks while
study area. To analyze the impact of COVID-19 pandemic on public using public transport. Respondents were asked to mention the fre­
transport usage, this study aimed to identify the factors at the beginning quency of travel by public transport before COVID-19 and foresight
of the pandemic that influenced the expected change in public transport about the expected change in the frequency of travel by public transport
usage during the ‘new normal’ situation. More specifically, it has two during an anticipated ‘new normal’ situation. The ‘new normal’ situa­
research objectives: 1) to examine the probable impact of COVID-19 on tion was explained to the respondents by presenting a hypothetical
travelers’ public transport usage during the ‘new normal’ situation, and scenario in the questionnaire. This hypothetical scenario was that until
2) to identify the effect of travelers’ characteristics, psychology, and the world has a proper cure for COVID-19, people need to adjust their
travel behavior, as well as pandemic-related factors on the expected lifestyle to fully start their activities by adopting preventive measures to
change in the frequency of travel by public transport during the ‘new protect themselves and others from COVID-19 infection. Compared to
normal’ situation. The study results would be valuable to combat issues the pre-COVID-19 situation, this new hypothetical situation was
with public transportation in any future pandemic, including any further considered as ‘new normal’ situation in this study. We constructed this
deterioration of the present COVID-19 condition. hypothetical scenario of ‘new normal’ situation based on reviewing
scares literatures available at that time. After data screening, all the
2. Methodology responses were found to be valid and complete. A comparison of the
socio-economic variables of the sample and the population is presented
2.1. First wave context in Table 1. Table 2 presents the details and descriptive statistics of the
variables considered for this study. Based on these two tables, sample
The first COVID-19 case in Bangladesh was detected on March 8, characteristics are described in the following paragraphs.
2020 (Zafri et al., 2021). As of March 20, 2022, about 470 million people The sample socio-economic characteristics suggest that 67% of the
have been infected by the virus, leading to 30 thousand deaths (GoB, respondents were male. In addition, most of the respondents were young
2022). On March 26, 2020, the country went under complete lockdown, (<30 years) and middle-aged (30–60 years) people. The average age of
and as a result, the offices, educational institutions, and shopping cen­ the respondents was 25.5 years. In a way, it helped our causes, as we
ters were closed (Anwari et al., 2021). After two months of closure, would have liked to focus on these groups because we had assumed that
almost everything except the educational institutions was gradually they were the main public transport user groups in Bangladesh. In
opened (Zafri et al., 2021a). In Bangladesh, public transport mainly general, limited numbers of females and a negligible number of older
implies the bus service within the cities as none of the cities has options people (60+ years) use public transport in Bangladesh due to poor

2
N.M. Zafri et al. Asian Transport Studies 9 (2023) 100099

Table 1 Table 2
Comparison of the socio-economic variables of the sample and the population. Summary statistics of the independent variables.
Variables Population Sample Statistics Variable name Variable Dummy Variable % Mean
description (Std.
Data Source Parameter
dev)
Total BBS (2017) 162.7 million 804
Socio-economic variables
Gender
Gender Gender of the If female, then 1; 32.96
Male BBS (2017) 50.03% 67.03%
respondent otherwise, 0
Female 49.97% 32.96%
Age Age of the Continuous 25.5
Age
respondent (5.7)
<30 Years BBS (2011) 62.16% 83.58%
Household Monthly household Low-income: If low- 19.77
30–60 years 30.37% 16.04%
income income of the income (<20000
>60 years 7.47% 00.37%
respondent in BDT BDT) then 1;
Monthly Household Income
otherwise 0
<20000 taka BBS (2010) 94.92% 19.78%
Middle-income: If 55.84
>20000 taka 5.08% 80.22%
middle-income
Household size BBS (2016) 4.06 4.55
(20000–60000
Living location type
BDT) then 1;
Divisional district BBS (2011) 26.61% 64.70%
otherwise 0
Others 73.39% 35.30%
High-income: If 24.37
BBS= Bangladesh Bureau of Statistics. high-income
(>60000 BDT) then
1; otherwise 0
service quality as well as user unfriendliness. Also, in an online survey, it Household size Number of Continuous 4.55
is not easy to incorporate older people, especially in developing coun­ members staying in (1.5)
tries, as they are not tech-savvy (Bhaduri et al., 2020). Consequently, the household of
the respondent
our survey was missing representation from the older group of people.
Location type Location type of If divisional district 64.70
In addition, people without internet access could not participate in the living place of then 1; otherwise 0
the survey. As a result, we found less participation from the low-income the respondent
group (monthly income less than <20000 BDT) and people who lived Travel behavior-related variables
outside divisional districts. Therefore, like many other studies con­ Car ownership Whether the If yes, then 1; 29.70
respondent’s otherwise, 0
ducted through the online platform during this pandemic period (Bha­ household owned
duri et al., 2020; Zafri et al., 2021; Anwari et al., 2021; Abdullah et al., any private car or
2020; Shakibaei et al., 2020), this study might also have some limita­ not.
tions related to sample attributes since our sample did not adequately Motorcycle Whether the If yes, then 1; 30.50
ownership respondent’s otherwise, 0
represent the population. Regardless of the limitations, sample charac­
household owned
teristics indicate that people from diverse spatial locations and any motorcycle or
socio-economic backgrounds participated in the survey, which might be not.
good enough to produce reliable results in the context of Bangladesh. Bicycle Whether the If yes, then 1; 32.60
In case of travel behavior-related variables in Table 2, around 29.7% ownership respondent’s otherwise, 0
household owned
of the respondents’ household owned at least one car; whereas, the any bicycle or not.
ownership percentages were 30.5% and 32.6% for motorcycle and bi­ Regular travel Regularly used Public mode: If 30.80
cycle, respectively. In case of regular mode usage during the pre-COVID- mode mode by the public transport,
19 period, respondents traveled through the public mode the most respondent during then 1; otherwise, 0
the pre-COVID-19 Shared mode: If 26.90
(30.8%), followed by shared mode (26.9%), private mode (22.8%), and
period shared mode
walking (19.5%). Compared to Dhaka Structure Plan (2016–2035), (CNGa, rickshawb,
private mode usage and ownership were found higher in the sample rideshare) then 1;
(RAJUK, 2015). The reason behind this was the presence of higher otherwise, 0
representation from high-income group in the sample. In addition, Private mode: If 22.80
private mode (car,
among the sample, around 40% of the respondents regularly used public motorcycle,
transport. Finally, 47% of the respondents expected to travel less during bicycle) then 1;
the ‘new normal’ situation as compared pre-COVID-19 situation; otherwise, 0
whereas, 23% expected to increase travel. Walk: If walk, then 19.50
1; otherwise, 0
In case of COVID-19 related variables (Table 2), respondents
Frequency of Frequency of travel Regularly: If 30.80
perceived high risk of COVID-19 transmission in public transportation. travel by by public transport regularly (>3 days
Previous studies also showed the similar result (Zafri et al., 2022; Bar­ public during the pre- per week) then 1;
bieri et al., 2021). In addition, respondents disagreed with the idea that transport pre- COVID-19 otherwise, 0
it might be possible to take preventive measures in public transport. COVID-19 pandemic period Usually: If usually 10.60
by the respondent (2–3 days per week)
Besides, around 75% of the respondents kept regular update about
then 1; otherwise, 0
COVID-19. Lastly, around 66% of the respondents expected to increase Occasionally: If 16.30
virtual activities (e.g., online class, office, shopping) in ‘new normal’ occasionally (at
situation. least one day per
week), then 1;
In case of psychology-related variables (Table 2), respondents had
otherwise, 0
strong trust on the preventive strategies, which suggested to keep one Rarely: If rarely, 42.30
safe from COVID-19 (4.13). Also, a large portion of the respondents then 1; otherwise, 0
concerned with the health and economic impact of the pandemic (4.06). Expected change Expected change in More: If more than 23.00
Few respondents felt that their immune system is strong enough and in trip the frequency of pre-COVID-19, then
frequency (all travel in the ‘new 1; otherwise, 0
they would not be affected by COVID-19 (2.35). Also, few respondents
(continued on next page)
wished to go outside social gatherings although there was risk of COVID-

3
N.M. Zafri et al. Asian Transport Studies 9 (2023) 100099

a
Table 2 (continued ) Three-wheeler motorized mode running on compressed natural gas (CNG)
and carrying 1–4 passengers.
Variable name Variable Dummy Variable % Mean b
description (Std.
A human paddled three-wheeler mode carrying 1–2 passengers.
dev)

modes) in the normal’ situation Same: If same as 30.00 19 transmission (1.85).

‘new normal’ compared to the pre-COVID-19, then
situation pre-pandemic 1; otherwise, 0
2.3. Data analysis
situation by the Less: If less than pre- 47.00
respondent COVID-19, then 1;
considering all otherwise, 0 Both descriptive analysis and statistical modeling approaches were
modes applied to analyze the data. Descriptive analysis was conducted to
COVID-19 related variables analyze the expected variation in public transport use before the COVID-
Perceived risk of Perceived risk of Five-point Likert 1.79
COVID-19 COVID-19 scale (1 = (1.2)
19 pandemic and during the ‘new normal’ situation. A multinomial lo­
transmission transmission in extremely high, 5 gistic regression model was developed to identify the factors influencing
public transport by = extremely low) the expected change in travel frequency by public transport in the ‘new
the respondent normal’ situation.
Possible to take Agreement with Five-point Likert 2.81
Multinomial logistic regression is one of the popular techniques used
preventive the statement: scale (1 = strongly (1.3)
measures “Preventive disagree, 5 = to analyze the association between a categorical dependent variable and
measures are strongly agree) multiple independent variables. It is also used to predict the dependent
possible to take in variable as the resultant outcome of the independent variable(s)
public transport." included in the model. In this method, the dependent variable should
Keeping updated Keeping Regularly: If 75.24
about COVID- themselves regularly (at least
have more than two nominal-discrete nature outcomes. If the dependent
19 updated about one time per day), variable has J outcomes, then J-1 logistic regression models will be
COVID-19 by the then 1; otherwise, 0 developed through this method (Kwak and Clayton-Matthews, 2002).
respondent Occasionally: If 11.56 Among the dependent variable outcomes, one outcome needs to be
occasionally (at
selected as a reference outcome to carry out the comparison with other
least one time per
week), then 1; outcomes. For example, the dependent variable has three outcomes (J =
otherwise, 0 3): x, y, and z, where x is considered the reference outcome. Hence, two
Rarely: If rarely, 13.18 models (J = 3–1 = 2) will be developed, where, in one model, y will be
then 1; otherwise, 0 predicted comparing with x, and in another one, z will be predicted
Expected change Expected change in More: If more than 66.00
in virtual virtual activities pre-COVID-19, then
comparing with x. Multinomial logistic regression can be generalized by
activities (online class, 1; otherwise, 0 the following equation, where J is the reference outcome.
office, shopping) in Same: If same as 17.80 [ ] [ ]
the ‘new normal’ pre-COVID-19, then πj P(y = j) ∑K
ln = ln = αj + βjk xk ; j = 1, 2, 3 J-1
situation compared 1; otherwise, 0 πJ P(y = J) k− 1
to the pre- Less: If less than pre- 16.20
pandemic situation COVID-19, then 1;
by the respondent otherwise, 0
Here, αj is the constant, and βjk is the coefficient of xk , an independent
Psychology-related variables variable for the jth outcome of the dependent variable.
Trust in Agreement with Five-point Likert 4.13 This study aimed to explore the expected change in travel frequency
preventive the statement: scale (1 = strongly (1.2) by public transport in the ‘new normal’ situation with three outcomes:
strategies “The more we take disagree, 5 =
less than pre-COVID-19, same as pre-COVID-19, and more than pre-
precautions and strongly agree)
follow the health COVID-19 situations. These three outcomes were considered as the
guidelines, the less three categories of the dependent variable in this study. The “Same as
is the risk of being pre-COVID-19′′ outcome was considered the reference category. Inde­
affected by COVID-
pendent variables considered to develop the models were organized
19."
Concern about Agreement with 4.06
under four broad categories: socio-economic variables, travel behavior-
COVID-19 the statement: “I (1.2) related variables, COVID-19 related variables, and psychology-related
impact am concerned variables. Table 2 presents the details and descriptive statistics of
about the health these independent variables considered for this study. Before developing
and economic
the model, we checked the strength of associations between the
impacts of COVID-
19 in Bangladesh." dependent variable and each of the independent variables. Cramer’s V
Feeling about the Agreement with 2.35 test was conducted to measure the strength of association with cate­
immune the statement: “I (1.1) gorical independent variables, whereas, for numeric independent vari­
system feel that my ables, we used the Eta Coefficient test. The strength of the association
immune system is
very strong, and I
varies from 0 to 1, where a value close to 0 indicates a weak relationship
am very less likely and a value close to 1 indicates a very strong association.
to be affected by
COVID-19." 3. Results and discussion
Willingness to go Agreement with 1.85
outside the statement: “If I (1.1)
am allowed, I am 3.1. Impact of COVID-19 on public transport usage
going to social
gatherings to meet Fig. 1 shows the frequency of travel by public transport before the
people even if
pandemic by the respondents. The Government of Bangladesh (GoB)
there is a risk of
COVID-19 imposed a countrywide lockdown from March 26, 2020 to May 31, 2020
transmission." (Anwari et al., 2021). Before the arrival of COVID-19 diseases, almost
20% of the respondents used public transport every day (Fig. 1). A
considerable portion of the respondents used public transport regularly

4
N.M. Zafri et al. Asian Transport Studies 9 (2023) 100099

Fig. 1. Frequency of travel by public transport before COVID-19.

(11% used 4–5 times a week, 11% used 2–3 times a week). Around 40% transport use during the ‘new normal’ situation, several studies suggest a
of the respondents rarely (23%) or never (18%) used public transport in reduction in the frequency of travel by public transport mode (De Vos,
the pre-COVID-19 situation. Similar trends are reported in previous 2020; Ipsos, 2020). However, surprisingly, there are still a good portion
studies as well. In Bangladesh, many people regularly used public of the respondents who were expecting to increase travel by public
transport for travel before the emergence of the COVID-19 pandemic, transport. A possible underlying reason could be their deteriorating
especially in urban areas. The trip share of public transport was around economic condition due to the pandemic (Bhuiyan et al., 2020).
37% in Dhaka (RAJUK, 2015).
The expected change in the frequency of travel by public transport of
3.2. Factors influencing expected change in frequency of travel by public
the respondents during the ‘new normal’ situation is presented in Fig. 2.
transport
Results show that around 60% of the respondents expected to travel less
by public transport during the ‘new normal’ situation than pre-
The results of both Cramer’s V test and Eta Coefficient test are pre­
pandemic. Furthermore, about 20% of the respondents did not expect
sented in Table 3. Ten independent variables were found to have a
any change in their frequency of travel by public transport. One of the
significant association with the expected change in the frequency of
most interesting and unexpected findings is that about 20% of the re­
travel by public transport in the ‘new normal’ situation. The strength of
spondents expected to increase travel by public transport (Fig. 2).
association showed that the dependent variable was strongly associated
As there was less representation from the low-income group in the
with the expected change in trip frequency in the ‘new normal’ situation.
sample, around 60% reduction in public transport usage might be a
In addition, household income, perceived risk of COVID-19 trans­
much-overestimated figure in a practical context. However, it is highly
mission, expected change in virtual activities, trust in preventive stra­
likely that a large portion of people might reduce travel by public
tegies, and concern about COVID-19 impact also had a slightly moderate
transportation. Public transport can be leveled as the transmission
association with the dependent variable (Table 3).
ground for viruses by bringing a non-infected passenger closer to an
A multinomial logistic regression model was developed with the
infected passenger (Tirachini and Cats, 2020; Troko et al., 2011).
significant independent variables to identify contributory factors that
Extreme overcrowding, cash payment of fare, human-to-human inter­
could influence the expected change in the frequency of travel by public
action, operation of improperly designed vehicles (e.g., having a single
transport in the ‘new normal’ situation. The Chi-square statistic for the
door, narrow gaps between two consecutive rows of seats), being un­
model was 742.373, which was found to be statistically significant at a
aware and lack of willingness to regularly disinfect vehicles are some of
99% confidence level (p < 0.000), indicating a good model fit (Table 4).
the main causes that increase the risk of the passengers being infected
Nagelkenke and McFadden pseudo-R-square values were found to be
even after taking proper personal hygiene measures (Islam, 2020; De
0.678 and 0.420, respectively, for the model, indicating that the inde­
Vos, 2020; Tirachini and Cats, 2020). Similarly, this study’s respondents
pendent variables could explain a good portion of the variation in the
perceived public transport as unsafe and expected to reduce travel by
expected change in the frequency of travel by public transport during
this mode compared to the pre-COVID-19 situation. Although there is a
the ‘new normal’ situation. The coefficient, p-value, and odds ratio
lack of research worldwide about the expected change in public
statistics for the significant independent variables are shown in Table 4.

Fig. 2. Expected change in the frequency of travel by public transport during the ‘new normal’ situation.

5
N.M. Zafri et al. Asian Transport Studies 9 (2023) 100099

Table 3 3.2.1. Socio-economic variable

Strength of association between the dependent variable and each of the inde­ Among the socio-economic variables, the respondent’s income was
pendent variables. only found to be statistically significant in the developed model. The
Variable name Cramer’s Eta (η) p-value result indicates that respondents from the middle-income groups were
V unlikely to increase their frequency of travel by public transport
Gender 0.055 0.308 (Table 4). The model results also indicate that high-income groups are
Age 0.069 0.107 more likely to travel less by using public transport in the ‘new normal’
Household income 0.109 0.001*** situation. These results indicate that while public transport might be an
Household size 0.007 0.852
unpopular option for middle-income and rich groups, low-income
Location type 0.077 0.092*
Car ownership 0.066 0.177 groups will usually keep using it the same as in pre-pandemic situa­
Motorcycle ownership 0.067 0.161 tions because of the unavailability of affordable and safer alternatives.
Bicycle ownership 0.097 0.023** The COVID-19 pandemic is devastating to the global economy as well as
Regular travel mode 0.099 0.048** the economy of Bangladesh. The World Bank estimated that the global
Frequency of travel by public transport pre- 0.088 0.053*
COVID-19
economy would shrink by 5.2% within 2020 (Maliszewska et al., 2020).
Expected change in trip frequency in ‘new 0.521 0.000*** In the case of Bangladesh, the GDP growth rate was estimated to reduce
normal’ from 8.0% to 2.0% in the financial year 2019-20 (IMF, 2020). Due to the
Perceived risk of COVID-19 transmission 0.218 0.000*** shutdown, lockdown, and the devastating impact of the pandemic on the
Keeping updated about COVID-19 0.061 0.200
economy, many people lost their jobs, so the income of a large number of
Expected change in virtual activities 0.173 0.000***
Possible to take preventive measures 0.038 0.117 people declined (Bodrud-Doza et al., 2020). A survey report showed that
Trust in preventive strategies 0.264 0.000*** poor people’s income dropped by 80% due to COVID-19 in Bangladesh
Concern about COVID-19 impact 0.225 0.000*** (Bhuiyan et al., 2020). As poor people become poorer, they are expected
Feeling about the immune system 0.065 0.111 to travel more in cheaper modes such as public transport regardless of
Willingness to go outside 0.015 0.471
the high risk of viral transmission associated with this mode, given the
***Significant at 99% confident level, ** Significant at 95% confident level,
*Significant at 90% confident level lack of options. Comparatively, middle- and high-income groups were
likely to travel less by public transportation as they have the financial
capabilities to shift to other safer modes.
Table 4
Results of multinomial logistic regression model. 3.2.2. Travel behavior-related variables
In the case of travel behavior-related variables, regular travel mode
Factors Expected change in frequency of travel by public transport
during pre-COVID-19, frequency of travel by public transport before
during ‘new normal’ situation (Ref: Same as pre-COVID-19)
COVID-19, and expected change in trip frequency in the ‘new normal’
Less than pre-COVID-19 More than pre-COVID-19
situation were found to be statistically significant. Regular private and
Coef. Odds p- Coef. Odds p- shared transport mode users are more likely to travel less by public
(B) Ratio value (B) Ratio value transport in the ‘new normal’ situation (Table 4). This indicates that
Constant − 1.952 0.142 0.000a − 2.226 0.108 0.000a public transport is becoming more unpopular among non-public trans­
Income port users due to the pandemic. This could also encourage a modal shift
Middle-income − 0.886 0.412 0.001a
from public transport towards private (e.g., car, motorcycle) and shared
High-income 0.532 1.702 0.061*
Regular travel mode modes (e.g., rickshaw, CNG, ridesharing). This potential modal shift
Private mode 0.773 2.166 0.010a might degrade traffic congestion, traffic safety, and environmental
Shared mode 0.819 2.268 0.002a condition and complicate the existing traffic management system. Non-
Frequency of public transport use before COVID-19 regular public transport users (those who mentioned their public
Usually − 0.708 0.493 0.049b
Rarely − 0.537 0.584 0.056*
transport use in the pre-pandemic situation as ‘Rarely’ or ‘Usually’) are
Expected change in overall trip frequency (by all modes) in the ‘new normal’ unlikely to travel less by public transport in the ‘new normal’ situation.
situation compared to the pre-pandemic situation As they are non-regular users, probably they are expecting no change in
More: Will make a 2.623 13.772 0.000a terms of their public transport use in the ‘new normal’ situation. On the
greater number
other hand, individuals who anticipated an increase in their overall trip
of trips
Less: Will make a 1.999 7.380 0.000a frequencies by all modes during the ‘new normal’ situation were more
less number of likely to increase their travel frequency by public transport. Similarly,
trips individuals who expected a decrease in their overall trip frequencies are
Perceived risk of 0.259 1.295 0.020b highly likely to decrease their travel frequency by public transport. The
COVID-19
transmission
plausible reason could be that the people who are prepared to travel
Expected change in virtual activities in the ‘new normal’ situation compared to more are likely to be less concerned about virus transmission risks on
the pre-pandemic situation buses than those who plan to limit their trips.
More: Will make a 0.521 1.684 0.016b
greater number
3.2.3. COVID-19 related variables
of virtual
activities Considering COVID-19 related factors, respondents who perceived a
Trust in 0.405 1.499 0.000a lower level of risk in public transport in terms of virus transmission were
preventive expected to increase travel by public transport during the ‘new normal’
strategies situation. Risk perception of people generally influences their risk-
Model statistics: χ2 = 742.373, df = 58, p-value = 0.000. taking behavior. People who perceive lower risk levels tend to engage
Nagelkenke ρ2 = 0.678, McFadden ρ2 = 0.420. in more risky behaviors and take less preventive measures to make them
a
Significant at 99% confident level. safe from exposure (Dryhurst et al., 2020; He et al., 2020). Therefore, the
b
Significant at 95% confident level, *Significant at 90% confident level. frequency of travel might increase if it is possible to lower people’s
perceived risk of public transport. In addition, respondents who ex­
pected to conduct more virtual activities during the ‘new normal’ situ­
ation were more likely to significantly reduce their travel by public

6
N.M. Zafri et al. Asian Transport Studies 9 (2023) 100099

transport. Due to the pandemic, more people have become used to vir­ issues, and advancements in technology differ from country to country,
tual activities such as online classes, working from home, online shop­ each country might need to adopt a unique set of policy responses
ping, mobile banking, online gathering and social activities, and online (Tirachini and Cats, 2020). The following ones are proposed for
food delivery services (Nicola et al., 2020), which might lead to reduced Bangladesh and similar developing countries, mainly its South Asian
travel demand for all modes including public transport. neighbors. These recommendations would help to combat COVID-19
and other pandemic-related adverse impacts in the future.
3.2.4. Psychology-related variables
Trust in preventive strategies was the only psychology-related vari­ • Our study indicates that risk perception associated with public
able that was significant in the model. The result indicates that a higher transport needs to be lowered, and trust in preventive strategies
level of trust in preventive strategies increased the likelihood of less needs to be established if it is aimed to raise public transport use to
travel by public transport during the ‘new normal’ situation. Studies pre-COVID-19 level or even more. Changes in the way public trans­
suggested that the probability of adopting preventive strategies would port services are operated should be made visible to the public. There
be higher if people had a higher level of trust in preventive strategies are instances of public transport drivers getting infected by passen­
(Dryhurst et al., 2020; Khosravi, 2020). People with a higher level of gers. It could be vice versa as well. Interaction between the public
trust in preventive strategies might understand the difficulties of prac­ transport service staff, including the drivers, and the passengers
ticing and maintaining these strategies in public transport. Therefore, should be minimized as much as possible. Tickets could be sold from
they expected to travel less by public transport in the ‘new normal’ temporary kiosks, set up in open spaces at the bus stop. Contagion
situation. risks are less in an open environment (Nishiura et al., 2020). Pas­
sengers standing in a queue and maintaining an adequate distance
4. Conclusion from each other could buy the ticket for the service they want. This
system is already in place for air-conditioned bus services in Dhaka.
This study examined the impact of the COVID-19 pandemic on public Contactless payment through BKash (a smartphone app), a method
transportation ridership during an anticipated ‘new normal’ situation in that has gained immense popularity lately, could be kept as an op­
Bangladesh. The study’s findings showed that 60% of the respondents tion. Only the passengers with a pre-purchased ticket would be
expected to reduce travel by public transport during the ‘new normal’ allowed onboard.
situation, whereas a small portion (20%) expected to increase. The • A significant portion of the respondents in our study still wanted to
collected sample for this study was overrepresented by middle- and use public transport in the ‘new normal’ situation, possibly due to
high-income groups (Table 1). This study also showed that the low- financial constraints or unavailability of alternatives. These passen­
income group would not reduce travel by public transport. If we gers are likely to feel comfortable if COVID-19 preventive protocols
generalized the result, it could be said that travel by public transport are applied inside the public transport vehicles. It is challenging to
would be reduced by middle- and high-income groups as they might ensure adequate physical distances among passengers on board. If
have better access to private modes. In general, these groups travel less some of the seats are left vacant, the operators need to be allowed to
by public transportation in Bangladesh due to a poor level of service. So, increase the fare to compensate for the loss in revenue. However, fare
overall, travel reduction by public transportation is likely to be less than increase is likely to discourage the passengers further from using
the estimated 60%. However, in countries where a large number of its public transport. These measures were already applied for about two
public transport ridership come from higher-income groups, travel months, but the operators and the passengers did not comply prop­
reduction is likely to be much higher. This would be a concerning issue erly (The Daily Star, 2020). All known occurrences of COVID-19
for those countries. infections in public transport vehicles happened when passengers
Besides income, respondent’s regular travel modes during the pre- were not wearing masks, including the one in Ningbo, China (Rabin,
COVID-19 period, their frequency of travel by public transport in the 2020). We strongly feel contagion probability can be significantly
pre-COVID-19 situation, their expected change in trip frequency and reduced if the passengers are mandated to wear masks properly. This
virtual activities during the ‘new normal’ situation, their risk perception, measure’s vigorous enforcement will restore people’s confidence,
and their trust in preventive strategies influenced the anticipated change especially the ones who perceive the transmission risk in public
of frequency of travel by public transport during the ‘new normal’ sit­ transport as a serious issue.
uation. In terms of effect sizes, the results indicate that the regular mode • To further reduce the chances of virus transmission on board and
used in the pre-pandemic period and the expected change in overall trip build up passengers’ trust, bus operators could follow strict hygiene
frequency by all modes in the ‘new normal’ situation are the most practices inside the bus by keeping the windows open to assist nat­
prominent factors determining public transport use in the ‘new normal’ ural ventilation and offering hand sanitizers to the passengers. After
situation. These results could offer useful insights for developing coun­ completing one way of each round trip, they would clean and
tries, especially in South Asia, with similar socio-economic, cultural, and disinfect the bus. There are only a few air-conditioned urban bus
geopolitical situations. services in Dhaka, which commenced only a couple of years ago.
From another perspective, although vaccine innovation reduced the These services should be kept suspended until the COVID-19 situa­
number of deaths due to COVID-19, it still cannot fully prevent people tion is no longer a threat to public health and safety.
from being infected by COVID-19. People are still being infected by • In order to reduce the perceived risk of the passengers, it is necessary
COVID-19. We can see a variety of COVID-19 variants emerging one to make the onboard environment even safer. Almost all the buses in
after another and creating a new wave of infection. In suchns, people Bangladesh have a single door. Otherwise, it would have been
have to carry out their daily activities by taking precautionary measures possible to arrange entry and exit through different doors. The op­
to save themselves from COVID-19, which resembles the ‘new normal’ erators should be encouraged to add two-door buses to their fleet in
situation we have defined. Therefore, apart from tackling new future the immediate future.
pandemics, the findings of the study are also likely to be useful for
policymakers in the prevailing situation. By adopting the appropriate policies, policymakers need to get the
As the results indicate a possible modal shift from public transport, confidence back in public transport users worldwide. If it is possible to
measures need to be taken to prevent this shift. Otherwise, private run the public transport services with minimum occurrences of COVID-
modes will overwhelmingly dominate mobility, hampering progress 19 infections linked to the services for a few months during high-
toward sustainable transportation outcomes and making equity objec­ infection periods, the confidence will further grow, and public trans­
tives unrealistic. As the intensity of the COVID-19 problems, contextual port is likely to capture back its modal share.

7
N.M. Zafri et al. Asian Transport Studies 9 (2023) 100099

References Mogaji, E., Adekunle, I., Aririguzoh, S., Oginni, A., 2022. Dealing with impact of COVID-
19 on transportation in a developing country: insights and policy recommendations.
Transport Pol. 116, 304–314. https://doi.org/10.1016/j.tranpol.2021.12.002.
Abdullah, M., Dias, C., Muley, D., Shahin, M., 2020. Exploring the impacts of COVID-19
Nicola, M., Alsafi, Z., Sohrabi, C., Kerwan, A., Al-Jabir, A., Iosifidis, C., et al., 2020. The
on travel behavior and mode preferences. Transp. Res. Interdiscip. Perspect. 8,
socio-economic implications of the coronavirus pandemic (COVID-19): a review. Int.
100255 https://doi.org/10.1016/j.trip.2020.100255.
J. Surg. 78, 185–193. https://doi.org/10.1016/j.ijsu.2020.04.018.
An, D., Tong, X., Liu, K., Chan, E.H.W., 2019. Understanding the impact of built
Nishiura, H., Oshitani, H., Kobayashi, T., Saito, T., Sunagawa, T., Matsui, T., et al., 2020.
environment on metro ridership using open source in Shanghai. Cities 93, 177–187.
Closed Environments Facilitate Secondary Transmission of Coronavirus Disease
https://doi.org/10.1016/j.cities.2019.05.013.
2019 (COVID-19). medRxiv.
Anwari, N., Tawkir Ahmed, M., Rakibul Islam, M., Hadiuzzaman, M., Amin, S., 2021.
Paul, T., Chakraborty, R., Afia Ratri, S., Debnath, M., 2022a. Impact of COVID-19 on
Exploring the travel behavior changes caused by the COVID-19 crisis: a case study
Mode Choice Behavior: A Case Study for Dhaka, Bangladesh. Transportation
for a developing country. Transp. Res. Interdiscip. Perspect. 9, 100334 https://doi.
Research Interdisciplinary Perspectives, 100665. https://doi.org/10.1016/j.
org/10.1016/j.trip.2021.100334.
trip.2022.100665.
Astroza, S., Tirachini, A., Hurtubia, R., Carrasco, J.A., Guevara, A., Munizaga, M., et al.,
Paul, T., Chakraborty, R., Anwari, N., 2022b. Impact of COVID-19 on daily travel
2020. Mobility changes, teleworking, and remote communication during the COVID-
behaviour: a literature review. Transport. Safety Environ. 4 (2), tdac013. https://
19 pandemic in Chile. Trasport Findings. https://doi.org/10.32866/001c.13489.
doi.org/10.1093/tse/tdac013.
Barbieri, D.M., Lou, B., Passavanti, M., Hui, C., Hoff, I., Lessa, D.A., et al., 2021. Impact
Paulley, N., Balcombe, R., Mackett, R., Titheridge, H., Preston, J., Wardman, M., et al.,
of COVID-19 pandemic on mobility in ten countries and associated perceived risk for
2006. The demand for public transport: the effects of fares, quality of service, income
all transport modes. PLoS One 16 (2), e0245886. https://doi.org/10.1371/journal.
and car ownership. Transport Pol. 13 (4), 295–306. https://doi.org/10.1016/j.
pone.0245886.
tranpol.2005.12.004.
Bhaduri, E., Manoj, B.S., Wadud, Z., Goswami, A.K., Choudhury, C.F., 2020. Modelling
Rabin, R.C., 2020. How a Bus Ride Turned into a Coronavirus Superspreader Event. https
the effects of COVID-19 on travel mode choice behaviour in India. Transp. Res.
://www.nytimes.com/2020/09/01/health/coronavirus-bus-china.html. (Accessed
Interdiscip. Perspect. 8, 100273 https://doi.org/10.1016/j.trip.2020.100273.
26 September 2020).
Bhuiyan, A.K.M.I., Sakib, N., Pakpour, A.H., Griffiths, M.D., Mamun, M.A., 2020. COVID-
RAJUK, 2015. Dhaka Structure Plan 2016-2035.
19-Related suicides in Bangladesh due to lockdown and economic factors: case study
Shakibaei, S., de Jong, G.C., Alpkökin, P., Rashidi, T.H., 2020. Impact of the COVID-19
evidence from media reports. Int. J. Ment. Health Addiction. https://doi.org/
Pandemic on Travel Behavior in Istanbul: A Panel Data Analysis. Sustainable Cities
10.1007/s11469-020-00307-y.
and Society, 102619. https://doi.org/10.1016/j.scs.2020.102619.
Bodrud-Doza, M., Shammi, M., Bahlman, L., Islam, A.R.M.T., Rahman, M.M., 2020.
Sogbe, E., 2021. The evolving impact of coronavirus (COVID-19) pandemic on public
Psychosocial and socio-economic crisis in Bangladesh due to COVID-19 pandemic: a
transportation in Ghana. Case Studies on Transport Policy 9 (4), 1607–1614. https://
perception-based assessment. [Original research]. Front. Public Health 8 (341).
doi.org/10.1016/j.cstp.2021.08.010.
https://doi.org/10.3389/fpubh.2020.00341.
The Daily Star, 2020. Public Transport: Pre-Covid Fare, Health Rules Not Followed.
De Vos, J., 2020. The effect of COVID-19 and subsequent social distancing on travel
Accessed 27 September 2020 2020. https://www.thedailystar.net/frontpage/new
behavior. Transp. Res. Interdiscip. Perspect. 5, 100121 https://doi.org/10.1016/j.
s/public-transport-pre-covid-fare-health-rules-not-followed-1954377.
trip.2020.100121.
Tirachini, A., Cats, O., 2020. COVID-19 and public transportation: current assessment,
Dryhurst, S., Schneider, C.R., Kerr, J., Freeman, A.L.J., Recchia, G., van der Bles, A.M.,
prospects, and research needs. J. Public Transport. 22 (1) https://doi.org/10.5038/
et al., 2020. Risk perceptions of COVID-19 around the world. J. Risk Res. 1–13.
2375-0901.22.1.1.
https://doi.org/10.1080/13669877.2020.1758193.
Troko, J., Myles, P., Gibson, J., Hashim, A., Enstone, J., Kingdon, S., et al., 2011. Is public
Gkiotsalitis, K., Cats, O., 2021. Public transport planning adaption under the COVID-19
transport a risk factor for acute respiratory infection? BMC Infect. Dis. 11 (1), 16.
pandemic crisis: literature review of research needs and directions. Transport Rev.
https://doi.org/10.1186/1471-2334-11-16.
41 (3), 374–392. https://doi.org/10.1080/01441647.2020.1857886.
UITP, 2020. Public Transport Authorities and COVID-19: Impact and Response to a
GoB, 2022. Latest Status in Bangladesh. https://corona.gov.bd/.
Pandemic. https://www.lek.com/sites/default/files/PDFs/COVID19-public-tr
He, S., Chen, S., Kong, L., Liu, W., 2020. Analysis of risk perceptions and related factors
ansport-impacts.pdf. (Accessed 30 May 2020).
concerning COVID-19 epidemic in chongqing, China. J. Community Health. https://
van Lierop, D., Badami, M.G., El-Geneidy, A.M., 2018. What influences satisfaction and
doi.org/10.1007/s10900-020-00870-4.
loyalty in public transport? A review of the literature. Transport Rev. 38 (1), 52–72.
Hensher, D.A., Beck, M.J., Wei, E., 2021. Working from home and its implications for
https://doi.org/10.1080/01441647.2017.1298683.
strategic transport modelling based on the early days of the COVID-19 pandemic.
van Soest, D., Tight, M.R., Rogers, C.D.F., 2020. Exploring the distances people walk to
Transport. Res. Pol. Pract. 148, 64–78. https://doi.org/10.1016/j.tra.2021.03.027.
access public transport. Transport Rev. 40 (2), 160–182. https://doi.org/10.1080/
IMF, 2020. Country News: Bangladesh. www.imf.org/en/Countries/BGD. (Accessed 15
01441647.2019.1575491.
August 2020).
Yezli, S., Khan, A., 2020. COVID-19 social distancing in the Kingdom of Saudi Arabia:
Ipsos, 2020. Impact of Coronavirus to New Car Purchase in China. https://www.ipsos.
bold measures in the face of political, economic, social and religious challenges.
com/en/impact-coronavirus-new-car-purchase-china.
Trav. Med. Infect. Dis., 101692 https://doi.org/10.1016/j.tmaid.2020.101692.
Islam, M.T., 2020. Social distancing in public transport yet another cruel joke. The
Zafri, N.M., Khan, A., Jamal, S., Alam, B.M., 2021. Impacts of the COVID-19 pandemic on
Business Standard. https://www.tbsnews.net/thoughts/social-distancing-public-tra
active travel mode choice in Bangladesh: a study from the perspective of
nsports-yet-another-cruel-joke-87106.
sustainability and new normal situation. Sustainability 13 (12). https://doi.org/
Jiang, F., Deng, L., Zhang, L., Cai, Y., Cheung, C.W., Xia, Z., 2020. Review of the clinical
10.3390/su13126975.
characteristics of coronavirus disease 2019 (COVID-19). J. Gen. Intern. Med. 35 (5),
Zafri, N.M., Afroj, S., Ali, M.A., Hasan, M.M.U., Rahman, M.H., 2021a. Effectiveness of
1545–1549. https://doi.org/10.1007/s11606-020-05762-w.
containment strategies and local cognition to control vehicular traffic volume in
Khosravi, M., 2020. Perceived risk of COVID-19 pandemic: the role of public worry and
Dhaka, Bangladesh during COVID-19 pandemic: use of Google Map based real-time
trust. Electron. J. General Med. 17 (4), em203.
traffic data. PLoS One 16 (5), e0252228. https://doi.org/10.1371/journal.
Kwak, C., Clayton-Matthews, A., 2002. Multinomial logistic regression. Nurs. Res. 51 (6).
pone.0252228.
Lipsitch, M., Swerdlow, D.L., Finelli, L., 2020. Defining the epidemiology of Covid-19 —
Zafri, N.M., Khan, A., Jamal, S., Alam, B.M., 2021b. Impact of COVID-19 pandemic on
Studies needed. N. Engl. J. Med. 382 (13), 1194–1196. https://doi.org/10.1056/
motorcycle purchase in Dhaka, Bangladesh. Front. Future Transport. https://doi.
NEJMp2002125.
org/10.3389/ffutr.2021.646664.
Luan, S., Yang, Q., Jiang, Z., Wang, W., 2021. Exploring the impact of COVID-19 on
Zafri, N.M., Khan, A., Jamal, S., Alam, B.M., 2022. Risk perceptions of COVID-19
individual’s travel mode choice in China. Transport Pol. 106, 271–280. https://doi.
transmission in different travel modes. Transp. Res. Interdiscip. Perspect. 13, 100548
org/10.1016/j.tranpol.2021.04.011.
https://doi.org/10.1016/j.trip.2022.100548.
Maliszewska, M., Mattoo, A., van der Mensbrugghe, D., 2020. The Potential Impact of
Zhang, R., Zhang, J., 2021. Long-term pathways to deep decarbonization of the transport
COVID-19 on GDP and Trade: A Preliminary Assessment (Policy Research Working
sector in the post-COVID world. Transport Pol. 110, 28–36. https://doi.org/
Papers). The World Bank.
10.1016/j.tranpol.2021.05.018.
Mogaji, E., 2020. Impact of COVID-19 on transportation in lagos, Nigeria. Transp. Res.
Interdiscip. Perspect. 6, 100154 https://doi.org/10.1016/j.trip.2020.100154.