Content uploaded by Ari K. M. Tarigan
Author content
All content in this area was uploaded by Ari K. M. Tarigan on Jul 10, 2019
Content may be subject to copyright.
Vol.:(0123456789)
Transportation
https://doi.org/10.1007/s11116-019-10019-5
1 3
To compete ornotcompete: exploring therelationships
betweenmotorcycle‑based ride‑sourcing, motorcycle taxis,
andpublic transport intheJakarta metropolitan area
MuhammadZudhyIrawan1· PrawiraFajarindraBelgiawan2·
AriKrisnaMawiraTarigan3· FajarWijanarko1
© Springer Science+Business Media, LLC, part of Springer Nature 2019
Abstract
In the last decade, the emergence of ride-sourcing services has transformed personal trip
behavior. In the context of Indonesia, ride-sourcing services have evolved into two modes
of transport: motorcycle-based and car-based. The presence of such services has strongly
impacted consumers’ choices of travel mode. However, the main question is whether the
ride-sourcing service is a complement or a substitute for the existing public transport and
conventional taxis. Using 438 motorcycle-based, ride-sourcing consumers, we applied a
structural equation model to investigate the relationships between motorcycle-based ride-
sourcing, motorcycle taxis, and public transport in the Jakarta Metropolitan Area. The
result shows that motorcycle-based ride-sourcing works as a complementary mode for
the TransJakarta bus and Jakarta commuter train, but as a competitor with the motorcy-
cle taxis. Contrarily, motorcycle taxis supported the existence of motorcycle-based ride-
sourcing. The study also found that individuals use motorcycle taxis as feeders to transit
stops. Individuals commonly use motorcycle taxis and motorcycle-based ride-sourcing for
short travel distances. The demographic features and technology use experience also drive
individuals’ choice of the three transportation modes. Integrating public transport with
motorcycle-based ride-source services, and legalizing motorcycle taxis and motorcycle-
based ride sourcing as forms of public transport are two main proposed policies that seek
to increase public transport demand, ensure service quality, safety, and fares, and reduce
the potential conflict between all three.
Keywords Motorcycle-based ride-sourcing· Public transport· Structural equation
modeling· Frequency of use
* Muhammad Zudhy Irawan
zudhyirawan@ugm.ac.id
1 Department ofCivil andEnvironmental Engineering, Universitas Gadjah Mada, Yogyakarta,
Indonesia
2 School ofBusiness andManagement, Institut Teknologi Bandung, Bandung, Indonesia
3 Department ofSafety, Economics, andPlanning, University ofStavanger, Stavanger, Norway
Transportation
1 3
Introduction
With the use of smartphone applications, on-demand ride-sourcing services have prolifer-
ated in the world during the past decade. Today, a user can simply request a ride through a
smartphone application, and the individual driver can respond in real time, leading to less
waiting and faster travel than public buses and even trains. Unlike traditional taxis, easier
calling for cars and easier payment lend ride-sourcing services a certain superiority.
However, ride-sourcing can be either complementary or substitutive toward public
transport and the conventional taxi (Rayle etal. 2016). While complementary refers to the
understanding that ride-sourcing may support the existing public transport service, substi-
tutive suggests that the latter may decrease the use of conventional public transport.
The rapid rise in ride-sourcing services has resulted in a substantial change in travel
behavior. Some studies have indicated that the high demand for ride-sourcing could poten-
tially reduce private vehicle usage and ownership (Metcalfe and Warburg 2012; Silver and
Fischer-Baum 2015), thus stabilizing auto ownership growth over the long term. On the
other hand, the technology has generated debate over its potential to undermine sustain-
able urban transportation. For example, although the emergence of ride-sourcing service
in major cities in the US could increase the commuter rail demand by 3%, it also threatens
to cut light rail and bus demand by reducing 3% and 6% respectively (Clewlow and Mishra
2017).
Congestion and environmental concerns are also potential downsides for ride-sourcing.
Henao (2017) shows that ride-sourcing increases vehicle miles traveled by 185 percent.
The cause is drivers circulating without riders, which has a significant impact on traffic
congestion and environmental issues. In addition, Rayle etal. (2016) shows that the use
of ridesharing services may induce people to travel, accounting for 8% of all trips. It is
because people have various desires and needs to travel from one point to another. The
adoption of ride-sourcing may maximize their ability to fulfill such needs and desires.
As the phenomenon of ride-sourcing has emerged, there remains scant empirical evi-
dence detailing the complex relationships between online and conventional public trans-
port services, especially in developing countries. Most of the studies investigating those are
focused on four-wheeled vehicles as the ride-sourcing mode (Dias etal. 2017; Rayle etal.
2016; Zha etal. 2017; Anderson 2014). Only a few studies have addressed travel behav-
ior among two-wheeled, or motorcycle-based, ride-sources (MBRS) (Santoso and Nelloh
2017; Silalahi et al. 2017; Medeiros et al. 2018). The main difference between the two
involves the utilization of app-based technology.
In developing countries, such as those in South East Asia, with poor public transport
services, motorcycle taxis have to an extent fulfilled that crucial role. Motorcycle taxis
act as an access/egress mode to buses and other transit hubs (see for example: Irawan and
Sumi 2011; Pongprasert and Kubota 2017; Idei and Kato 2019). Unlike four-wheeled taxis,
however, motorcycle taxis operate with unclear regulation, control, and management.
The two motorcycle modes also fill gaps in the public transport network service. Medei-
ros etal. (2018) show that 24% of commuter train users and 3% of bus rapid transit users
ride either motorcycle taxis or MBRS as their access/egress mode to and from bus or tran-
sit stations because of poor public transport service coverage in Jakarta. It is still unclear if
and how MBRS affect customer use of motorcycle taxis to public transport services.
Sukor etal. (2018) emphasize the urgency of conducting a campaign to repopularize
the public bus. They suggest offering monetary incentives. With increased traffic, however,
adopting ride-sourcing services could slow the process. Thus, this study may be the first
Transportation
1 3
effort to examine the relationship between online and offline transportation services, and
the various forms of public transport modes in Indonesia, and throughout South East Asia.
In particular, this study seems to be the first to uncover structural relationships related to
travel behaviors among patrons of MBRS, motorcycle taxis, and other public transport.
The issue seems to be poorly understood, to date, especially compared to similar cases in
industrialized countries.
Our objectives in this study were twofold: first, to determine how public transport,
motorcycle taxi, and MBRS affect each other. Do they complement or replace each other,
or neither, or both? Second, we examine how socio-demographics, Information and Com-
munication Technology (ICT) usage (e.g., smartphone apps), and travel behavior influence
customer decisions to use MBRS. Using data collected from MBRS users in the Jakarta
Metropolitan Area (JMA) in 2016, we performed structural equation modeling (SEM) with
the frequency of use and its perceived benefits as endogenous variables, and the factors of
travel time, the ICT experience, and socio-demographics as the exogenous variables.
In this paper, this section is followed by a literature review. Section3 describes the his-
tory of ride-sourcing in Indonesia, particularly in the JMA. In Sect.4, we discuss the sur-
vey design and research methodology. In Sect.5, we present the result of the SEM, fol-
lowed by a discussion in Sect.6. Finally, in the last section, we conclude our findings.
Literature review
Over the past few decades, the complex and hidden relationships between ICT and trans-
portation have been widely studied (Salomon 1986; Senbil and Kitamura 2003; Choo
etal. 2007). There are many types of ICT, with each form having a heterogeneous effect
on travel behavior. Types of ICT include e-shopping (Cao etal. 2012; Farag etal. 2007;
Irawan and Wirza 2015), telework (Melo and Silva 2017; Helminen and Ristimäki, 2007),
video conferencing (Denstadli 2004), real-time bus arrival information system (Brakewood
etal. 2015; Watkins etal. 2011), and ride-sourcing (Rayle etal. 2016; Anderson 2014).
The forms of ICT-transportation interaction can be classified in four ways: substitution,
modification, complementation, and neutrality (Senbil and Kitamura 2003). Substitution
means that with greater ICT use, the number of trips and/or activity duration decrease.
That’s the opposite of complementation, where the number of trips or activity duration
increase. Modification and neutrality mean that the use of ICT modifies aspects of the
trips and/or activities, such as the routes and timing of trips, and the timing, location and
sequence of activities. This is relevant to app-based motorcycle ride-sourcing, since the use
of ICT complements the number of trips or activity duration.
To date, several studies on the impacts of ride-sourcing on different aspects of trans-
portation have been conducted. Rayle et al. (2016) analyzed usage characteristics and
potential impacts of ridesharing and online taxis on travel behavior in San Francisco. Dias
etal. (2017) investigated the effects of socioeconomic and demographic variables on the
frequency of ride-source and rideshare use. Other studies explored user satisfaction and
MBRS service quality (Silalahi etal. 2017; Santoso and Nelloh 2017). Hall etal. (2018)
conducted a study on whether ride-sourcing (in this case Uber) complements or substitutes
for public transit. They found that ride-sourcing is a complement to small transit agencies
and to transit agencies in large cities. Feigon and Murphy (2016) distributed an online sur-
vey to approximately 4500 commuters to explore travel behavior and mode choice in rela-
tion to shared mode and transit use. They found that ride-sourcing services complemented
Transportation
1 3
public transit, especially during off-peak hours, where transit runs infrequently or not at all.
The two studies seem to suggest that ride-sourcing is a complement to the current existing
public transit network.
Furthermore, Chen (2015) found that of all ride-sourcing respondents in Pittsburgh, 82%
of them used ride-sourcing for social/leisure trips and only 15% for work. That study sug-
gests that ride-sourcing users are likely to switch from private vehicle use to ride-sourcing
mostly for social and leisure travels. Concerning the influencing variables of ride-sourcing
choices, the study found that convenience, high speed, and low cost are the primary rea-
sons the respondents chose them. In contrast, Dawes (2016) demonstrated that “safety” and
“comfort” are determining factors when choosing ride-sourcing, compared to conventional
public transport modes. It seems that users were uncertain about the maintenance proce-
dures for the vehicle and the skill of the ride-sourcing drivers to drive safely, compared
to the standard of conventional public transport services. According to Dawes (2016),
most of the ride-sourcing non-users said that the reason for not choosing ride-sourcing
were because it was expensive, unsafe, inconvenient, and slow. Some users thought that
one or two ride-sourcing companies benefited from the lack of regulatory oversight, which
unfairly favored the conventional taxis and the consumers.
According to some studies, socio-demographic attributes such as age, gender, income,
and educational level were basic determinants for the adoption of ride-sourcing. Rayle
etal. (2016) showed that ride-sourcing users in San Francisco were largely young (73% of
them less than 35years old) and well-educated (84% of them held bachelor’s degrees or
higher). Caulfield (2009) further demonstrated that females and younger individuals were
more likely to use ride share, whereas those 35–44years old were the least likely. Using a
bivariate ordered probit model, Dias etal. (2017) found that older individuals and lower-
income people had lower propensities to use ride-sourcing services, whereas individuals
with higher education levels were more likely to use them. Note the overlap between those
target groups. Regarding motorcycle taxis, in Bangkok, middle level income groups tend
to use cheaper motorcycle taxis to go to bus and train stations compared to lower-income
users. Lower-income groups prefer to walk (Pongprasert and Kubota 2017). In Vietnam,
lower-to-middle income groups tend to use motorcycle taxis for shopping, going to work,
study or leisure compared to higher income groups (Tuan and Mateo-Babiano 2013).
Those studies reported that monthly income is one variable that significantly correlates
with mode choice. However, this remains unclear in the case of MBRS.
Note also that people who use ride-sourcing are more likely to use it for short trips
(Rayle etal. 2016; Chen 2015; Tuffour and Appiagyei 2014). Rayle etal. (2016) found that
San Francisco commuters can save 33min of travel time by using public transport, but the
trip can additionally be shortened by 22min by using ride-sourcing. Related to the variable
of ICT, Dias etal. (2017) found that smartphone ownership produced a greater possibility
to use ride-sourcing. Smartphone ownership and use have also substantial effect on the
number of trip chains and the use of multiple transport means (Astroza etal. 2017).
MBRS inJMA
Since early 2015, online transport services of both four- and two-wheeled taxis have
become increasingly popular in Indonesia. This is supported by the fact that Indonesia is
one of the countries with the highest mobile and smartphone ownership rate in the world
(Ministry of Communication and Information 2009). Currently, more than 154 million
Transportation
1 3
mobile phones have been purchased in Indonesia, whereas the total population in the coun-
try is about 250 million people. Uber, Grab, and Go-Jek have the largest market shares
of the ride-sourcing service in Indonesia (Silalahi et al. 2017). The share of users of this
service has grown rapidly. For example, since it was launched on December 2014 until
the beginning of 2019, Go-Jek, the most popular ride-sourcing operator in Indonesia, has
reported that its app has been downloaded more than 50 million times. In addition, the Go-
Jek operator claims that over two million people have joined as motorcycle drivers serving,
in 50 cities across Indonesia.1
The two-wheeled vehicle is an informal public transport in Indonesia. In fact, the motor-
cycle taxi and MBRS are illegal, in terms of national2 and JMA3 regional regulations.
The absence of legal enforcement affects the price, safety, and customer protection. The
motorcycle taxi is individually operated by the driver. The tariff is determined by bargain-
ing between driver and passenger (Wicaksono etal. 2015). Meanwhile, the MBRS tariff
is determined by the company unilaterally (Roberto 2016). At the end of 2015, the Indo-
nesian Ministry of Transportation banned MBRS operation in the country. The regulation
was met with enormous public resistance. The public asked why is the ban only intended
for MBRS and not for motorcycle taxis? This debate compelled the Ministry of Transpor-
tation to allow the operation of MBRS services in the same year that the system had been
prohibited from operating.4 To date, more than three years later, there has been no change
in the law for the legitimacy of both MBRS and motorcycle taxi.
Our study was conducted in the JMA, the capital city of Indonesia and the largest met-
ropolitan area in South East Asia. Although covering only 0.034% of the total land in Indo-
nesia, JMA’s GDP is approximately 14% of the national GDP in 2016 (Central Bureau
of Statistic 2017). In the JMA, users suffer chronic traffic congestion for their daily trips.
Congestion in the urban areas of Indonesia has worsened over the past few years, owing to
the increase in private vehicle ownership and use (Joewono etal. 2016). There are about 26
million trips per day in the JMA, of which approximately 19 million are within the JMA,
and the remaining 7 million from its border areas: Bekasi in the east, Depok and Bogor in
the south, and Tangerang in the west. Nearly 98% of commuting trips outside the JMA are
made by private vehicles (Central Bureau of Statistic 2015a). JUTPI (2012) reported that
the motorcycle is the most widely used vehicle within the JMA with a share of 53%, fol-
lowed by private cars with 20%, and public transport with 27 percent. However, there is
no specific information about the percentage of public transport use of TransJakarta BRT,
commuter lines, and paratransit.
Travel patterns are complex in the JMA owing to the large amount of people
with different origins and destinations (Yagi and Mohammadian 2010). Accord-
ing to the dataset of the study for the Integrated Transportation Master Plan for
Jakarta–Bogor–Depok–Tangerang–Bekasi, in 2002 the number of trips and trip chains for
people living in the JMA was 2.7 and 1.2 per day, respectively (SITRAMP Jabodetabek
2004). That means in average each person having 2–3 trips a day. Also, each person on
1 See: www.go-jek.com/go-ride/.
2 See the Act No. 22/2009 (Traffic and Road Transport) Article No. 138. Available at: http://hubda t.dephu
b.go.id/uu/288-uu-nomor -22-tahun -2009-tenta ng-lalu-linta s-dan-angku tan-jalan .
3 See Regional Regulation of Jakarta Province No. 5/2014. Available at: https ://pelay anan.jakar ta.go.id/
downl oad/regul asi/perat uran-daera h-nomor -5-tahun -2014-tenta ng-trans porta si.pdf.
4 See: https ://www.theja karta post.com/news/2015/12/18/jokow i-defen ds-ride-haili ng-apps-trans porta tion-
minis try-withd raws-ban.html.
Transportation
1 3
average performed at least one trip chain and at most two trip chains a day. An example of
one trip chain for shopping is house-shopping mall-house again.
The use of MBRS is central to this study, first and foremost, because of the motorcy-
cle’s role supporting daily activities in the JMA and its fringe areas, especially among low-
income individuals (Senbil et al. 2006). The World Bank (2015) stated that 10% of the
population in Indonesia was living in poverty and receiving less than 2 USD/day. In the
JMA itself, the ownership of motorcycles reached 13 million in 2014, 75% of all motor-
ized vehicles, increasing 11% per year (Central Bureau of Statistic 2015b). The motorcy-
cle is very popular in developing countries because riders can easily maneuver in narrow,
crowded, and shabby roads (Sukor etal. 2017).
That said, even as the motorcycle has become a highly useful means of transport,
implicit in its value is its considerable cost to the society as a result of traffic injuries and
deaths (Vasconcellos 2013). In Indonesia, the number of traffic accidents shows a rising
trend, growing 16.59% per year. The Ministry of Transportation (2015) recorded 95,906
traffic accidents in 2014, in which 28,297 victims died, and 26,840 victims suffered severe
injuries. Out of that number, 108,883 motorcycles were involved in traffic accidents. The
first number refers to the number of accidents. Each accident might involve zero, one, two
motorcycles, or even more. Therefore, assuming that each accident involves two motor-
cycles, there will be at least 191,812 motorcycles involved. Thus, the number still makes
sense.
Second, the existence of MBRS is arguably helpful to support community activities.
The services provided by MBRS are not limited to public transportation. They can also be
used for goods and food delivery services (Wirawan and Oktivera 2015). Before the exist-
ence of MBRS, people were motivated to use the motorcycle taxi rather than the public
transport service because of the savings in time obtained. The motorcycle taxi can bring
the user exactly to the place where he/she wants to go, and this mode is often associated
with shorter travel times (Hagen etal. 2016). Travel time is also one of the determinants
that make train commuters prefer the motorcycle taxi as their access/egress mode (Irawan
etal. 2017).
It should be pointed out that the public transport referred to in this study is only the
TransJakarta bus rapid transit and commuter railway. The TransJakarta network length is
around 251.2km with 260 bus stops in 13 corridors. The price of TransJakarta ticket is
around IDR 20005 (5 to 7 AM); IDR 3500 (7 AM to 12 PM); and IDR 3500 (12 PM to 5
AM). The commuter railway, on the other hand, has a route length of 418.5km with 79 sta-
tions. The price of the commuter railway depends on the distance between IDR 3000–IDR
8000.
Research design andmethods
Survey design anddata collection
We designed a travel behavior questionnaire to understand the interactions between pub-
lic transport, motorcycle taxis, and MBRS usage. The questionnaire was categorized in
five sections. The first section asked about the respondents’ trip frequency when using
5 IDR 1000 = USD 0.071.
Transportation
1 3
MBRS, motorcycle taxi, and/or public transport over the previous week. In the second sec-
tion, the respondents were required to identify their last trip on a ride-sourcing application
to record their points of origin and destination and travel time. This was followed by a
question about the trip’s purpose. The third section is related to the perceived benefits of
choosing the MBRS, motorcycle taxi, and/or public transport on a five-point Likert scale,
from “completely disagree” to “completely agree.” Even though the Likert scale has been
widely used, there is debate about its application (Jamieson 2004; Hodge and Gillespie,
2003). Likert scales are frequently used in transportation studies, especially for construct-
ing latent variables. For example, Sohn and Yun (2009) applied Likert scales for deter-
mining the latent variables reflecting a car-dependent commuter’s psychometric factors by
using explanatory factor analysis. These latent variables are then considered influencing
factors on the mode choice model, in addition to the respondents’ characteristics and travel
mode performance. Farag etal. (2007) used the Likert scale to measure the attitudinal fac-
tors of online and in-store shopping as latent variables. They found that attitudes related to
e-shopping have an indirect effect on in-store shopping frequency.
In this study, we also implemented Likert scales for understanding the psychological
reasons why a traveler might use one or another mode. The fourth section asked about their
experiences with ICT, such as how long they have used ride-sourcing applications and their
experience in ordering taxis (four- or two-wheelers) by phone before the ride-sourcing ser-
vice was available. The questionnaire closed by asking for socio-demographic characteris-
tics, such as age, gender, education level, and income.
The interview survey was conducted from March to July 2016. To maintain the accu-
racy of the collected data, we involved well-trained surveyors who were familiar with the
JMA. Both surveyors and respondents were compensated.
Surveyors began by showing respondents their official survey permission from the
Transportation Research and Development Agency, of the Indonesian Ministry of Trans-
portation. They asked respondents to consent to the study and whether they fulfilled the
initial requirements. The first requirement was that the respondent should have used the
three transport modes over the previous week. The second was that the respondents should
work and perform daily activities in the JMA, though they may come from outside the
area (Bogor, Depok, Tangerang, Bekasi). Surveyors also informed the respondents that the
collected data was used for research purposes only, and that there would be no questions
regarding their name, address, phone number, or email address. Surveyors went to random
key locations in Central Jakarta and its vicinity that were expected to have a high concen-
tration of MBRS users, such as railway stations, bus stops, shopping areas, and workplaces.
Those were chosen because most of the public transport hubs, shopping centers, and gov-
ernment offices were concentrated there. Surveyors randomly intercepted MBRS users on
the street in key locations. Hours of the survey were divided into two time blocks—7 to 10
AM and 4 to 7 PM. “Appendix 1” shows the flowchart outlining the criteria and sampling
progress.
To aid the surveyor in collecting the data, we involved two types of respondents: users
who had just completed a MBRS trip and those who had used MBRS within the last week.
For both respondent types, they had to show their trips recorded on their smartphone
applications.
Note that there were two main possible shortcomings to the data, and consequent limita-
tions to the study. First, even though in the data collection process we had carefully selected
the respondents so that they reflected the JMA population data, the data collected did not
necessarily represent the JMA population that used public transport, motorcycle taxi, and
MBRS for daily trips. Second, although the face-to-face interview process was carefully
Transportation
1 3
carried out, we were not able to obtain comprehensive information from the MBRS user
because of the respondents’ limited time, given that we interviewed them on the street and
in locations with high concentrations of MBRS users.
Description ofthesample
Out of the 786 respondents who participated in the survey, 438 (55.72%) completed the
questionnaire. Of those 438, 396 (90.41%) made trips within the JMA, while 42 (9.59%)
had at least an origin or destination elsewhere.
From Table1, we could see that the MBRS survey respondents primarily fell within the
age range of 15–24years old (71%) and were largely well-educated (66% of MBRS users
held bachelor’s degrees or higher). Around 68% of the MBRS respondents were women.6
A similar result was also demonstrated by Medeiros etal. in their 2018 study. There, they
revealed that 65% of MBRS users in JMA were female. In addition, Iskandar etal. (2017)
showed that 58% of MBRS users in JMA are women, with 60% of them younger than
30years old. With respect to their incomes, because the majority of respondents were less
than 24years old, almost all MBRS users had low and medium incomes, 43% and 41% of
them, respectively. Since the proportion of the socio-demographic categories did not match
the population proportions, this study considers the sampling weight method to have nor-
malized the data using “post-stratified weights,” which is similar to what Belgiawan etal.
(2019) did with Jakarta data. The sampling weight was then used for the model estimation.
We also found that most (46%) used MBRS less than three times in a week. Only 18%
used public transport no more than three times per week. The majority (45%) used public
transit three-to-six times per week; indeed, one-third of the MBRS customers used it more
than seven times in a week.
This mode use frequency phenomenon suggests that MBRS seems to act as a primary
access for public transport use. This assertion is strengthened by the survey result of
MBRS use showing that the service is generally used for short trips and, in most cases, the
travel time was either less than 20min (42%) or between 20 and 40min (35%). Our survey
results in Table2 also show that in the case of trips to and from transit points, the question
arises whether MBRS will support the public transport service. Twenty-point-three per-
cent and 14.6% of MBRS users ride it to reach and return from a railway station/bus stop,
respectively. Meanwhile, 6.2% of them ride it to transfer between public transport modes,
as, for example, from a bus stop to a railway station.
Although MBRS use potentially complements public transport use, it is also highly
possible that it substitutes for the use of public transport, as our survey result reveals. In
that case, 311 MBRS users (71.2%) made direct trips from origin to destination. Of those,
42.2% started their trips from home. The highest percentage of MBRS users’ trip purposes,
by 32.4%, was a trip to the workplace, while the smallest percentage of MBRS users’ trip
purposes was for shopping activity. The MBRS users also tended to abandon the motorcy-
cle taxi, since our survey result revealed that 84% of MBRS users chose it less than three
times in a week and none of them used motorcycle taxi more than seven times in a week.
6 Survey respondents were generally younger, better educated, and more represented by women than the
average population in the JMA. Ten respondents were younger than 15years old, specifically 13 to 14year
old junior high school students. It was feared that their responses might bias the results. However, since
no respondents were in elementary school and they were able to travel without parental supervision, we
decided to include this data.
Transportation
1 3
Table 1 Descriptive statistics of endogenous and exogenous variables
Variable N % Mean SD JMAa
Key variables
Frequency of public transport use
1. Less than 3 times/week 81 18
2. 3–6 times/week 196 45
3. 7–10 times/week 93 21
4. 11–14 times/week 51 12
5. More than 14 times/week 17 4
Frequency of MBRS use
1. Less than 3 times/week 200 46
2. 3–6 times/week 197 16
3. 7–10 times/week 31 7
4. 11–14 times/week 10 2
5. More than 14 times/week 0 0
Frequency of motorcycle taxi use
1. Less than 3 times/week 368 84
2. 3–6 times/week 70 16
3. 7–10 times/week 0 0
4. 11–14 times/week 0 0
5. More than 14 times/week 0 0
Socio-demographics
Gender
1. Male 142 32 51
2. Female 296 68 49
Age
1. Less than 15years old 10 2 25
2. 15–24years old 309 71 16
3. 25–34years old 73 17 21
4. 35–44years old 37 8 17
5. More than 44years old 9 2 22
Education
1. Elementary school 0 0 22
2. High school 148 34 68
3. Academic degree and higher 290 66 10
Income per month
1. Low income (less than IDR 2.5 million)* 187 43
2. Medium income (IDR 2.5 to 6 million)* 181 41
3. High income (more than IDR 6 million)* 70 16
ICT usage experience
Ride-sourcing application use duration
1. Less than 7months 199 45
2. 7–2months 147 34
3. 1–1.5years 58 13
4. 1.5–2years 23 5
5. More than 2years 11 3
Transportation
1 3
Regarding their ICT experience, the majority of respondents (79%) has used a ride-
sourcing application for less than a year. A total of 36% of the respondents reported that
they had no experience ordering a taxi, either a two-wheeler vehicle or four-wheeler vehi-
cle, before the ride-sourcing service was available, and only 12% and 13% of the respond-
ents ordered taxi several times and 1–2 times in a month, respectively. Out of the respond-
ents who had never ordered a taxi before, 29% are people between 15 and 24years old,
while the rest (7%) are people more than 24years old. It seems that young people were
more likely to have never ordered a taxi before the proliferation of ride-sourcing apps.
SEM analysis
We employed SEM to analyze the correlation between MBRS, motorcycle taxis, and pub-
lic transport. SEM is a multivariate regression that is commonly used in travel behavior
research (Golob 2003). For instance, Kim and Goulias (2004) used SEM to measure the
relationship between activity, travel, and telecommunications technology in their report to
the annual meeting of the Transportation Research Board. By using SEM, a variable could
a Source: JMA in Figure (in %), 2016, *IDR 1000 = USD 0.071
Table 1 (continued)
Variable N % Mean SD JMAa
Ordering experience via telephone before ride-sourcing application is available
1. Never 159 36
2. 1–2 times in a year 65 15
3. Several times in a year 104 24
4. 1–2 times in a month 57 13
5. Several times in a month 53 12
Travel time
1. Less than 20min 183 42
2. 20–40min 153 35
3. 40–60min 84 19
4. 1–1.5h 7 2
5. More than 1.5h 11 3
Utility variable
The perceived benefits of public transport use
Effective and convenience for long distance 4.11 0.76
Safe 4.08 0.83
Comfort (such as protected from weather) 4.18 0.74
Support the government policy in reducing congestion 4.31 0.76
The perceived benefits of MBRS use
Short waiting time 4.28 0.79
No walking 4.66 0.54
Easy to use 4.41 0.67
The perceived benefits of motorcycle taxi use
Fast 4.47 0.64
Reliable 4.33 0.74
Flexible 4.50 0.63
Transportation
1 3
Table 2 Trip origin and destination of MBRS users
Destination Total
Home School Transit points Workplace Social/Leisure Shopping area Other % n
Origin Home – 6.8 6.2 21.5 8.0 3.9 2.3 48.6 213
School 1.8 0.9 2.7 1.1 1.4 0.5 0.2 8.7 38
Transit points 2.5 2.3 6.2 2.1 1.1 – 0.5 14.6 64
Workplace 3.2 0.5 2.3 4.3 3.0 – 0.9 14.2 62
Social/Leisure 1.1 – 0.7 0.2 0.2 0.2 – 2.5 11
Shopping area 3.2 – 1.1 2.7 0.9 – 0.2 8.2 36
Other 0.9 0.2 1.1 0.5 – – 0.5 3.2 14
Total % 12.8 10.7 20.3 32.4 14.6 4.6 4.6 100 438
n 56 47 89 142 64 20 20 438
Transportation
1 3
play as an outcome variable and an explanatory variable simultaneously. In addition, the
correlation produced by SEM distinguishes between direct and indirect effects. A SEM
analysis consists of a measurement model and a structural model. In the first instance, the
observed variables construct the latent variables. The second model results in the correla-
tion between endogenous (dependent) variables, and between endogenous and exogenous
(independent) variables.
We determined six endogenous variables in a model that consisted of the frequency of
public transport, motorcycle taxis, and MBRS use, and the perceived benefits obtained by
using public transport, motorcycle taxi, and MBRS. We argue that reciprocal relationships
exist between the variables of MBRS usage, motorcycle taxi usage, and public transport
usage. This means that (1) the use of MBRS may affect the uses of a motorcycle taxi and
public transport, (2) the use of the motorcycle taxi may influence the uses of MBRS and
public transport, and (3) the use of public transport may influence the uses of MBRS and
motorcycle taxi. We also suspect that the perceived benefits of using the three modes are
likely to influence people’s decisions to choose a transport mode. Lastly, we assume that
socio-demographics, ICT usage experience, and travel time may be determinants that influ-
ence mode choice.
It should be clear that the variables of the perceived benefits from using a given mode
are considered latent. The endogenous variables are influenced by several exogenous vari-
ables, such as travel time, an experience of ICT use, and socio-demographics. Figure 1
illustrates the proposed relationships between endogenous variables and between endog-
enous and exogenous variables.
Fig. 1 Proposed hypothesized SEM model
Transportation
1 3
Results
After we construct and run the hypothesized model (Fig. 1), we need to test the multi-
variate normality of variables. The result shows that the multivariate kurtosis value is
0.985 and the critical value is 0.358. As the critical value is below 1.96, we find that our
model signifies the multivariate normality with 95% confidence. Next, we check the model
fit results. According to Golob (2003), there are several “goodness of fit” indices that are
frequently used to assess the SEM output, including the maximum likelihood ratio Chi
square, Hoelter critical N, root mean square error of approximation (RMSEA), and good-
ness of fit index (GFI).7 The Chi square value/df is 4.13, the Hoelter critical N is 145, the
GFI is 0.915, and the RMSEA is 0.08. As all indices are a good or reasonable fit, it can be
concluded that the model performs reasonably well.
We included latent variables in our model to accommodate the users’ perception related
to the service offered by public transport, motorcycle taxi, and MBRS. To construct the
latent variable, we have to fix one of its observed indicators on one (Hox and Bechger
1998). Looking at the t-statistics value in Table3, all observed indicators for the latent
variables performed well.
In our proposed model, we assume that public transport, motorcycle taxi, and MBRS
usages have caused a reciprocal relationship (as shown by the six links in the dashed box
in Fig.1). We also tested whether the trip frequency of the different modes was influenced
by the perceived benefits of using the given mode. The relationships between variables
are reported in Table4 for the direct and total effect. The relationship between key vari-
ables, the frequency of using public transport, the frequency of using motorcycle taxi, and
frequency of using MBRS are significant when compared with each other. As expected,
we found a significant negative effect of MBRS use on motorcycle taxi use. We also found
Table 3 Latent construct
Latent variables Parameter estimate t-statistics
The perceived benefits of public transport use
Effective and convenience for long distance 1.000
Safe 1.465 8.844
Comfort (such as protected from weather) 1.882 8.423
Support the government policy in reducing congestion 1.832 8.558
The perceived benefits of MBRS use
Short waiting time 1.000
No walking 1.276 7.652
Easy to use 0.650 5.479
The perceived benefits of motorcycle taxi use
Fast 1.000
Reliable 0.244 4.581
Flexible 0.449 5.537
7 Good fit if GFI ≥ 0.9; a critical Hoelter critical N of 200 or better indicates a satisfactory fit, while a value
under 75 is unacceptable; good fit if RMSEA ≤ 0.05, and reasonable fit if RMSEA ≤ 0.08; good fit if Chi
square/df ≤ 3, and adequate fit if chi-square/df ≤ 5.
Transportation
1 3
Table 4 Direct and total effect (p < 0.01)
Explanatory variables Dependent variables
Frequency of public
transport use
Frequency of motorcy-
cle taxi use
Frequency of MBRS
use
Perceived benefits of
public transport use
Perceived benefits
of motorcycle taxi
use
Perceived benefits of
MBRS use
Endogenous variables
Key variables
Frequency of public
transport use
0.000 (− 0.057*) 0.369* (0.312*)
Frequency of motorcy-
cle taxi use
0.267** (0.145**) 0.905* (0.542*)
Frequency of MBRS
use
− 0.911** (− 0.492*)
Mode choice behavior
Perceived benefits of
public transport usea
0.348 (0.343) 0.000 (− 0.02*)
Perceived benefits of
motorcycle taxi usea
0.000 (0.066*) 0.202** (0.248) − 0.283** (− 0.034)
Perceived benefits of
MBRS usea
0.166* (0.084*)
Exogenous variables
Trip characteristic:
travel time
− 0.109** (− 0.018**) − 0.134** (− 0.119**)
Experience of use technology
Duration of ride-sourc-
ing application use
− 0.139 (− 0.167) − 0.148* (− 0.105*) 0.000 (− 0.089*)
Taxi-ordering experi-
ence by phone
− 0.105* (− 0.104*)
Transportation
1 3
(1) An empty cell means the coefficient is insignificant at the 0.10 level (2) The numbers in parentheses are total effects, (3) a direct effect of 0.000 means the coefficient for
the direct link between variables is insignificant at the 0.10 level, (4) ameans latent variable, (5) *means p < 0.05, (6) **means 0.05 ≤ p < 0.1
Table 4 (continued)
Explanatory variables Dependent variables
Frequency of public
transport use
Frequency of motorcy-
cle taxi use
Frequency of MBRS
use
Perceived benefits of
public transport use
Perceived benefits
of motorcycle taxi
use
Perceived benefits of
MBRS use
Socio-demographic variables
Age 0.067* (0.067*) 0.052** (0.052**)
Gender
Education 0.000 (0.091**) − 0.171* (− 0.067*) − 0.089** (− 0.089**) 0.209 (0.209)
Income 0.000 (− 0.129*) − 0.215* (− 0.266*) 0.346* (0.03) − 0.103* (− 0.103*)
Transportation
1 3
that some latent variables of mode preferences influenced the frequency of mode use. We
discuss the SEM results in more detail in the next section.
Discussion
As depicted in Fig.2, our result shows that the frequency of motorcycle taxi usage has a
positive impact on the use of MBRS. In contrast, negative significant effects occur from
MBRS on the use of motorcycle taxis, meaning that individuals who have a greater trip fre-
quency on MBRS may be less likely to use motorcycle taxis. It appears that while the pres-
ence of MBRS is seen as a competitor for the motorcycle taxi, the motorcycle taxi might be
seen as a supporter of the MBRS.
The estimated model further suggests that motorcycle taxi use may positively and sig-
nificantly imply the use of public transport, meaning that individuals who often use motor-
cycle taxis may be likely to frequently use public transport. Traditionally, motorcycle taxis
provide service for short distance travel with quite an expensive fare. Therefore, motorcy-
cle taxis are mostly used as a feeder from home to the nearest public transport network.
On the other hand, there is a different relationship between MBRS usage and public
transport usage. The estimated model reports that the use of MBRS seems to have no
impact on public transit use. It can be argued that individuals who are using MBRS are
not public transport users. It can be also interpreted that there is a mode shift from private
vehicle modes to MBRS usage. However, individuals who often use public transport may
be likely to use MBRS frequently. It means that public transport users tend to use MBRS
for trips from a transit stop. In this case, MBRS acts as a complement for public transport.
Thus, individuals who using MBRS partly also public transport users. Moreover, the result
also demonstrates that, indirectly, the use of public transport may decrease the likelihood
Fig. 2 SEM results
Transportation
1 3
of motorcycle taxi usage. This is consistent with our earlier findings demonstrating that
MBRS acts as a competitor of motorcycle taxi, especially on trips from transit stop.
As expected, the perceived benefits of public transport usage positively affect the fre-
quency of use of public transport. When individuals feel a safe and comfortable service,
as well as a more effective and convenient mode for long distance trips, it increases the
frequency of public transport use. A similar relationship between perceived benefits and
frequency occurs for both motorcycle taxi and MBRS. It shows that the higher the utility
obtained by the user of transport means, the higher the frequency of using that mode. A
short waiting time, easy to use, and no need walking to find MBRS are key factors in deter-
mining people’s preference of MBRS. However, the benefits perceived by motorcycle taxi
are likely to decrease the use of MBRS. It means that the possibility that a rider utilizes
MBRS will decrease when the motorcycle taxi offers a more flexible and reliable service,
for example, in a condition of the high demand for MBRS, peak hours, or when it is raining
causing a much longer waiting time, very expensive tariff, and even order rejection by the
MBRS driver.
In terms of the relationship between endogenous and exogenous variables, as shown in
Fig.2, we found that the longer the travel time, the fewer the number of trips made using
motorcycle taxi and MBRS. This might indicate that the motorcycle taxi and MBRS are
commonly used for short travel distances, especially when it performs in filling gaps in
public transport service coverage. On the other hand, MBRS could also be a substitute for
a longer trip since mode shifting is happening from private vehicle modes to MBRS usage.
With respect to the variable of experience with technology usage, taxi-ordering experi-
ence by phone has a negative effect on public transport usage. It has an insignificant effect
on trip-making using motorcycle taxi and MBRS. Travelers who commonly order taxis via
phone call will cut their public transport usage. Subsequently, duration of usage of ride-
sourcing applications may have direct negative effects on the frequency of public trans-
port and motorcycle taxi usage. People who have used a ride-sourcing application on their
smartphone for a long time tend to use less public transport and motorcycle taxis. How-
ever, out of our expectation, the duration of usage of ride-sourcing applications indirectly
affects, negatively, the frequency of MBRS usage. We assume that they have understood
when they have to use or leave MBRS, owing to the fluctuating tariff and waiting time of
the MBRS service at peak and off-peak hours, as well as at low and high demand seasons.
Therefore, it seems that the use of MBRS decreases.
Considering the variable of socio-demographics, income directly affects the frequency
of motorcycle taxi and MBRS usage and decreases the use of public transport mode indi-
rectly. The positive effect of income on the frequency of MBRS usage and the negative
effect on the frequency of motorcycle taxi usage shows that people with higher income
more often use MBRS and tend to lessen the usage of motorcycle taxis. The low service
standard of motorcycle taxis makes it less preferable over the MBRS for individuals with
high income. It is clear that the adoption of MBRS is basically an answer to improve the
quality of services that are substantially missing from motorcycle taxi. Unlike motorcycle
taxis, MBRS users can give a rating score for the drivers associated with their service to
customers. Medeiros etal. (2018) shows that more than 80% of MBRS users are satisfied
with MBRS service, while only 48% of motorcycle taxi users are satisfied with its service.
By using MBRS, people do not need to negotiate tariffs with drivers. Customers can also
call the driver anywhere and anytime, whereas for motorcycle taxi customers have to walk
to find it first or maybe waiting for it. Although the higher income users more often use
the MBRS mode, they are less likely to perceive its benefits. Most likely, they tend to have
a higher dependency on the MBRS mode for their daily trips, even though they feel less
Transportation
1 3
benefit from that mode. In addition, their dependency on MBRS is not only influenced by
the decrease in motorcycle taxi usage as its competitor mode but it also indirectly causes a
decrease in the frequency of public transport usage.
Hereinafter, despite our expectation, the estimated model indicates that education has
direct and negative effects on MBRS. This result is contrary to the context of income,
showing from the estimated model that individuals with higher income are likely to use
MBRS. How and why individuals with a higher degree of education are less likely to use
MBRS are difficult to explain in this research. Thus, this might remain as further studies.
We propose the future analysis will first recode the education variable into the interaction
variable that considers both education and income aspects. This recoded explanatory vari-
able may provide a better picture to explain the effect of the education-income variable to
the likelihood of using MBRS. The education level also has an insignificant effect on the
frequency of public transport use, representing that there is no effect of education on public
transport usage. In addition, more educated people perceive a disadvantage in the use of
public transport and a benefit in MBRS use. The disadvantage of public transport use is
related to the currently unreliable public transport service, while the benefit of the MBRS
is related to its role in filling the gap of public transport service.
This study also found that age has an insignificant effect on the use of public transport,
MBRS, and motorcycle taxis. However, as the age factor has positive coefficients on the
dependent variables of perceived benefits of public transport and MBRS use, it reveals that
older people realize the benefits provided by public transport and MBRS if they have to
choose these two modes for their trips. Lastly, in terms of gender, it has an insignificant
influence on public transport, motorcycle taxi, and MBRS choice decision.
Conclusion
This paper contributes to the current knowledge on travel behavior with respect to the ride-
sourcing service. The study identifies the complex relationships across the uses of MBRS,
motorcycle taxi, and public transport. Our study was conducted among respondents in the
JMA based on an SEM framework. To the best knowledge of the authors, this is the first
attempt in research that examines the hidden influences of MBRS services on the uses of
other public transport forms.
Our result indicates that competition exists between motorcycle taxis and MBRS in
the JMA. The analysis reported that the use of MBRS might reduce the use of motorcy-
cle taxis. Because motorcycle taxis are considered one of the backbones of the informal
transport mode in JMA, the presence of MBRS may result in their diminishing role. This
might cause uncertainty in terms of income and future role among the drivers of motorcy-
cle taxis. Medeiros etal. (2018) show that the daily income of MBRS drivers is nearly 1.7
higher than motorcycle taxi drivers (USD 12.8 vs. 7.6). It then remains a question for the
policymakers whether they simply leave this competition to the market or provide a policy
package to enable the motorcycle taxi to compete with the MBRS in the search of potential
passengers. However, we recommend that the government regulate both the motorcycle
taxi and MBRS and consider the motorcycle taxi and MBRS as forms of public transport
with specific requirements. As we have discussed in the background section, because the
Act No 22/2009, Article No. 138 does not consider two-wheeled vehicles as legitimate
public transport, the government cannot yet do so.
Transportation
1 3
Because of this, MBRS and motorcycle taxis remain unlawful modes of travel. Once
the motorcycle taxi and MBRS are included in the act, the government can easily regulate
them, not only with regard to minimum standards of service but also with regard to safety.
By so doing, the government could also control the numbers of drivers of both types and
determine their operating zones to reduce the potential conflict between them.
Our result also reveals a positive effect of the use of public transport on MBRS use.
An increase in public transport implies an increase in MBRS usage. This demonstrates
that MBRS could support a public transport service and offer an optimistic insight for
policymakers and scholars about the positive relationship between the two modes in the
future. Policymakers, who in this case are the JMA authorities, might focus their attention
on improving the maintenance of the public transport, as we found that the MBRS could
support the existence of public transport. The authorities should also integrate the public
transport service and MBRS services, for example, by integrating a ticketing system and
then giving an integrated ticket discount, as well as providing an integrated service applica-
tion. A lesson can be drawn from The Southeastern Pennsylvania Transportation Authority
(SEPTA)—Uber rideshare partnership. SEPTA decided to work together with Uber aiming
to increase their train demand by giving 40% discount for Uber rides with a maximum dis-
count of 10 USD per ride. As a consequence of providing subsidies that are entirely borne
by Uber, SEPTA provided co-branded advertising on their railcars.8
Concerning the effect of travel time, we found that individuals who faced longer travel
times tended to take fewer trips, whether by MBRS or motorcycle taxi. This demonstrates
the central role of both modes in supporting users with a shorter travel times. The study
also found that people with higher income levels were more likely to use MBRS than
motorcycle taxis. The SEM result further indicates that gender and age have no correla-
tion with the frequency of MBRS, motorcycle taxi, and public transport use. As the level
of education has a negative impact on the frequency of MBRS use, the study might indi-
cate that the MBRS is more popular among less-educated individuals. Lastly, related to the
duration of ride-sourcing application use, although nearly 80% of the MBRS consumers
installed the ride-sourcing application on their smartphone less than a year ago, this fac-
tor has a negative effect on their frequency to use public transport and the motorcycle taxi.
This factor also has a negative and indirect effect on the frequency of MBRS use. Perhaps,
MBRS consumers are aware of the ride-sourcing characteristics, and thus, they can easily
decide on what condition they would use MBRS or choose the alternative modes.
We suggest that future studies explore these hypotheses using longitudinal data. This
would provide a much more detailed analysis regarding how such relationships change
over a span of time. Future studies also need to expand the analysis of the other form of
ride-sourcing, the car taxi. The impact of ride-sourcing on paratransit mode is also impor-
tant to be examined. In addition, it might be fruitful to consider whether temperature con-
straints, weather, and traffic congestion have influenced the uses of MBRS. As the adoption
of MBRS is still new in Indonesia, a future study may be needed to investigate the role of
MBRS in other metropolitan areas of Indonesia, such as Bandung, Surabaya, and Medan.
This study has focused on travel behavior aspects among different forms of public trans-
port mode. It may be essential to elaborate the structural organization for operating the
whole public transport system in the city. This issue may help verify the critical points
that limit the reform process of the JMA public transport system. Related to the limitation
8 See: https ://www.isept aphil ly.com/blog/Uber.
Transportation
1 3
of this study, future work should collect a large sample and comprehensive information of
ride-sourcing users. The data cleaning process and sample bias correction process must be
carefully executed. The SEM model could be further refined by adding more independent
variables.
Acknowledgements This research has been supported by the Indonesian Ministry of Research, Technology,
and Higher Education. We are grateful for the anonymous reviewers whose comments helped to improve the
paper. The remaining errors are our own (Grant No. 170/UN1/DITLIT/DIT-LIT/LT/2018).
Compliance with ethical standards
Conict of interest The corresponding author states that there is no conflict of interest.
Appendix1
Respondent recruitment and questionnaire survey flow chart.
Transportation
1 3
References
Anderson, D.N.: Not just a taxi”? For-profit ridesharing, driver strategies, and VMT. Transportation
41(5), 1099–1117 (2014)
Astroza, S., Garikapati, V.M., Bhat, C.R., Pendyala, R.M., Lavieri, P.S., Dias, F.F.: Analysis of the
impact of technology use on multimodality and activity travel characteristics. J. Transp. Res. Board
2666, 19–28 (2017)
Belgiawan, P.F., Ilahi, A., Axhausen, K.W.: Influence of pricing on mode choice decision in Jakarta: a
random regret minimization model. Case Stud. Transp. Policy 7(1), 87–95 (2019)
Brakewood, C., Macfarlane, G.S., Watkins, K.: The impact of real-time information on bus ridership in
New York City. Transp. Res. Part C: Emerg. Technol. 53, 59–75 (2015)
Cao, X.J., Xu, Z., Doum, F.: The interactions between e-shopping and traditional in-store shopping: an
application of structural equations model. Transportation 39, 957–974 (2012)
Caulfield, B.: Estimating the environmental benefits of ride-sharing: a case study of Dublin. Transp. Res.
Part D Transp. Environ. 14(7), 527–531 (2009)
Central Bureau of Statistic: Indonesia Economic Report. https ://www.bps.go.id/index .php/publi kasi/
index (2017). Accessed 8 July 2018
Central Bureau of Statistic: Commuting Trip in JMA 2014. https ://jakar ta.bps.go.id/backe nd/brs_ind/
brsIn d-20150 22009 4832 (2015a). Accessed 10 July 2018
Central Bureau of Statistic: Transportation Statistics of JMA 2015. http://jakar ta.bps.go.id/backe nd/pdf_
publi kasi/Stati stik-Trans porta si-DKI-Jakar ta-2015.pdf. (2015b). Accessed 10 July 2018
Chen, Z.: Impact of Ride-Sourcing Services on Travel Habits and Transportation Planning. Master of
Science Thesis. University of Pittsburgh, Pittsburgh, Pennsylvania (2015)
Choo, S., Lee, T., Mokhtarian, P.: Do transportation and communications tend to be substitutes, com-
plements, or neither? US consumer expenditures perspective, 1984-2002. Transp. Plan. Technol.
30(5), 431–453 (2007)
Clewlow, R.R., Mishra, G.S.: Disruptive transportation: The adoption, utilization, and impacts of ride-
sourcing services in San Francisco. https ://itspu bs.ucdav is.edu/wp-conte nt/theme s/ucdav is/pubs/
downl oad_pdf.php?id=2752 (2017). Accessed 28 Feb 2019
Dawes, M.:Perspectives on the Ridesourcing Revolution: Surveying Individual Attitudes Toward Uber
and Lyft to Inform Urban Transportation Policymaking. Ph.D. Thesis. Massachusetts Institute of
Technology, Cambridge, Massachusetts (2016)
Denstadli, J.M.: Impacts of videoconferencing on business travel: the Norwegian experience. J. Air
Transp. Manage. 10(6), 371–376 (2004)
Dias, F.F., Lavieri, P.S., Garikapati, V.M., Astroza, S., Pendyala, R.M., Bhat, C.R.: A behavioral choice
model of the use of car-sharing and ride-sourcing services. Transportation 44(6), 1307–1323 (2017)
Farag, S., Schwanen, T., Dijst, M., Faber, J.: Shopping online and/or in-store? A structural equation
model of the relationships between e-shopping and in-store shopping. Transp. Res. Part A Policy
Pract. 41(2), 125–141 (2007)
Feigon, S., Murphy, C.: Shared mobility and the transformation of public transit. https ://www.apta.com/
resou rces/repor tsand publi catio ns/Docum ents/APTA-Share d-Mobil ity.pdf (2016). Accessed 11 Mar
2019
Golob, T.F.: Structural equation modeling for travel behavior research. Transp. Res. Part B Methodol.
37, 1–25 (2003)
Hagen, J.X., Pardo, C., Valente, J.B.: Motivations for motorcycle use for Urban travel in Latin America:
a qualitative study. Transp. Policy 49, 93–104 (2016)
Hall, J.D., Palsson, C., Price, J.: Is Uber a substitute or complement for public transit? Journal of Urban
Economics 108, 36–50 (2018)
Helminen, V., Ristimäki, M.: Relationships between commuting distance, frequency and telework in
Finland. J. Transp. Geogr. 15(5), 331–342 (2007)
Henao, A.: Impacts of ridesourcing–lyft and uber–on transportation including vmt, mode replacement,
parking, and travel behavior (2017). http://digit al.aurar ia.edu/conte nt/AA/00/00/60/55/00001 /
Henao _ucden ver_0765D _10823 .pdf. Accessed 6 June 2019
Hodge, D.R., Gillespie, D.: Phrase Completions: an alternative to Likert 17 scales. Soc. Work Res. 27,
45–55 (2003)
Hox, J.J., Bechger, T.M.: An introduction to structural equation modeling. Family Sci. Rev. 11, 354–373
(1998)
Idei, R., Kato, H.: Medical-purposed travel behaviors in rural areas in developing countries: a case study
in rural Cambodia. Transportation, 1–24. (2019)
Transportation
1 3
Iskandar, H., Wibowo, I., Subagja, I.K.: Effect of customer value and quality of service on customer sat-
isfaction (Case study on consumers GO-JEK, Jakarta Indonesia). Int. J. Adv. Sci. Res. 2(5), 28–39
(2017)
Irawan, M.Z., Putri, M.K., Belgiawan, P.F., Dwitasari, R.: Analyzing commuters’ behavior on egress trip
from railway stations in Yogyakarta Indonesia. Open Transp. J. 11, 1–15 (2017)
Irawan, M.Z., Sumi, T.: Modeling departure time of students on public transports in Yogyakarta Indone-
sia. Mod. Appl. Sci. 5(4), 3–10 (2011)
Irawan, M.Z., Wirza, E.: Understanding the effect of online shopping behavior on shopping travel
demand through structural equation modeling. J. Eastern Asia Soc. Transp. Stud. 11, 614–625
(2015)
Jamieson, S.: Likert scales: how to (ab)use them. Med. Educ. 38, 1217–1218 (2004)
Joewono, T.B., Tarigan, A.K.M., Susilo, Y.O.: Road-based public transportation in urban areas of Indone-
sia: what policies do users expect to improve the service quality? Transp. Policy 49, 114–124 (2016)
JUTPI: Jabodetabek Urban Transportation Policy Integration Project. Ministry of Economic Affairs of
the Republic of Indonesia & Japan International Cooperation Agency, Jakarta (2012)
Kim, T.G., Goulias, K.G: Cross sectional and longitudinal relationships among information and telecommu-
nication technologies, daily time allocation to activity and travel, and modal split using structural equa-
tion modeling. In83rd Annual Meeting of Transportation Research Board, Washington, DC (2004)
Medeiros, R.M., Duarte, F., Achmad, F., Jalali, A.: Merging ICT and informal transport in Jakarta’s ojek
system. Transp. Plan. Technol. 41(3), 336–352 (2018)
Melo, P.C., Silva, J.D.A.: Home telework and household commuting patterns in Great Britain. Transp.
Res. Part A Policy Pract. 103, 1–24 (2017)
Metcalfe, G., Warburg, J.: A Policy agenda for the sharing economy. The Urbanist (2012)
Ministry of Communication and Information: Statistics of Mobile Phone Users, Jakarta. http://stati stik.
komin fo.go.id/ (2009). Accessed 25 Aug 2018
Ministry of Transportation: Land Transport in Figure 2014, Jakarta. http://hubda t.dephu b.go.id/data-
a-infor masi/pdda/tahun -2015/1765-perhu bunga n-darat -dalam -angka -edisi -april -2015 (2015).
Accessed 4 Aug 2018
Pongprasert, P., Kubota, H.: Switching from motorcycle taxi to walking: A case study of transit station
access in Bangkok, Thailand.IATSS Research (2017)
Rayle, L., Dai, D., Chan, N., Cervero, R., Shaheen, S.: Just a better taxi? A survey-based comparison of
taxis, transit, and ridesourcing services in San Francisco. Transp. Policy 45, 168–178 (2016)
Roberto, I.: The phenomena of motorcycle-taxi or ojek as public transportation in Jakarta and its legiti-
macy according to traffic and transportation law. Problematika Hukum 2(1), 35–45 (2016)
Salomon, I.: Telecommunications and travel relationships: a review. Transp. Res. Part A General 20(3),
223–238 (1986)
Santoso, A.S., Nelloh, L.A.M.: User satisfaction and intention to use peer-to-peer online transportation:
a replication study. Procedia Comput. Sci. 124, 379–387 (2017)
Senbil, M., Kitamura, R.: The Use of Telecommunications Devices and Individual Activities Relation-
ships. Paper Presented at the Transportation Research Board (TRB) 82nd Annual Meeting, Wash-
ington, D.C., U.S.A (2003)
Senbil, M., Zhang, J., Fujiwara, A.:Motorcycle Ownership and Use in Jabotabek (Indonesia) Metro-
politan Area. http://home.hiros hima-u.ac.jp/~hicec /coe/produ cts/DP200 6/DP200 6-3.pdf (2006).
Accessed 26 July 2018
Silalahi, S.L.B.R., Handayani, P.W., Munajat, Q.: Service quality analysis for online transportation ser-
vice: case study of GO-JEK. Procedia Comput. Sci. 124, 487–495 (2017)
Silver, N., Fischer-Baum, R.: Public Transit Should Be Uber’s New Best Friend.FiveThirtyEight (2015)
SITRAMP Jabodetabek: A Study of Integrated Transportation Master Plan for Jakarta-Bogor-Depok-
Tangerang-Bekasi 2002 [Database] (2004)
Sohn, K., Yun, J.: Separation of car dependent commuters from normal choice riders in mode choice
analysis. Transportation 36, 423–436 (2009)
Sukor, N.S.A., Basri, N.K., Tarigan, A.K.M.: The role of incentives towards adolescents’ commitment to
use public transport in Malaysia. Transportation Planning and Technology. 1–18 (2018)
Sukor, N.S.A., Tarigan, A.K.M., Fujii, S.: Analysis of correlations between psychological factors and
self-reported behavior of motorcyclists in Malaysia, depending on self-reported usage of different
types of motorcycle facility. Transp. Res. Part F Traffic Psychol. Behav. 46, 509–523 (2017)
Tuan, V.A., Mateo-Babeano, I.B.: Motorcycle taxi service in Vietnam: its socioeconomic impacts and
policy considerations. J. Eastern Asia Soc. Transp. Stud. 10, 13–28 (2013)
Tuffour, Y.A., Appiagyei, D.K.N.: Motorcycle taxis in public transportation services within the Accra
Metropolis. Am. J. Civil Eng. 2(4), 117–122 (2014)
Transportation
1 3
Vasconcellos, E.A.: Road safety impacts of the motorcycle in Brazil. Int. J. Inj. Control Saf. Promot. 20(2),
144–151 (2013)
Watkins, K.E., Ferris, B., Borning, A., Rutherford, G.S., Layton, D.: Where is my bus? Impact of mobile
real-time information on the perceived and actual wait time of transit riders. Transp. Res. Part A Policy
Pract. 45(8), 839–848 (2011)
Wicaksono, A., Lim, I., Muromachi, Y., Vergel, K.N., Choocharukul, K., Tan, V.H., Terai, K., Fukuda, D.,
Yai, T.: Road-based urban public transport and paratransit in Six Asian Countries: legal conditions and
intermodal issues. J. Eastern Asia Soc. Transp. Stud. 11, 227–242 (2015)
Wirawan, F.W., Oktivera, E.: Analysis on the implementation of digital marketing towards motorbike trans-
port service case study: Go-Jek (online taxi motorbike) Jakarta, Indonesia. International Conference on
Information Technology Systems and Innovation (ICITSI), November 16-19, 2015. Bandung Institute
of Technology, Bandung, Indonesia (2015)
World Bank: World Development Indicators: Poverty Rates at International Poverty Lines. http://wdi.world
bank.org/table /2.8 (2015) Accessed 24 June 2018
Yagi, S., Mohammadian, A.K.: An activity-based microsimulation model of travel demand in the Jakarta
metropolitan area. J. Choice Model. 3(1), 32–57 (2010)
Zha, L., Yin, Y., Du, Y.: Surge pricing and labor supply in the ride-sourcing market. Transp. Res. Procedia
23, 2–21 (2017)
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Muhammad Zudhy Irawan is an assistant professor of transportation planning and modeling in the Depart-
ment of Civil and Environmental Engineering, Universitas Gadjah Mada, Indonesia. His research interests
are related to travel behavior, modeling and forecasting, public transport, and intelligent transport system.
He also develops a ZIN macro simulation software and smartphone-based bus tracking system for Trans-
Jogja Bus in Yogyakarta, Indonesia.
Prawira Fajarindra Belgiawan is a lecturer in the School of Business and Management, Institut Teknologi
Bandung, Indonesia. He earned his M.Eng. degree in 2012 and his Ph.D. degree in 2015 from the Kyoto
University, Japan. He was also a postdoctoral researcher in the Institute for Transport Planning and Systems,
ETH Zurich, Switzerland from 2016 to 2017. His research activities focus on choice modeling, travel behav-
ior, demand models, and network models.
Ari Krisna Mawira Tarigan is an associate professor in the Department of Safety, Economics and Planning,
University of Stavanger, Norway. He earned his Ph.D. degree in Transportation from the Kyoto University,
Japan, and his M.Sc. degree from the IHE Delft, Netherlands. His research interests include road safety, traf-
fic engineering, travel behavior, transportation planning, and urban infrastructure.
Fajar Wijanarko is a supporting staff. He received his Bachelor’s degree in Civil and Environmental Engi-
neering from the Universitas Gadjah Mada, Indonesia.
A preview of this full-text is provided by Springer Nature.
Content available from Transportation
This content is subject to copyright. Terms and conditions apply.
