Content uploaded by Nirajan Shiwakoti
Author content
All content in this area was uploaded by Nirajan Shiwakoti on Jun 10, 2024
Content may be subject to copyright.
Research in Transportation Business & Management 56 (2024) 101147
Available online 4 June 2024
2210-5395/© 2024 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Integrating perceived safety and socio-demographic factors in UTAUT
model to explore Australians' intention to use fully automated vehicles
Yilun Chen
a
,
b
, Shah Khalid Khan
a
,
c
, Nirajan Shiwakoti
a
,
*
, Peter Stasinopoulos
a
,
Kayvan Aghabayk
d
a
School of Engineering, RMIT University, Melbourne, Australia
b
Simulation and Modelling Team, Transport for New South Wales, Sydney, Australia
c
Center of Cyber Security Research & Innovation, RMIT University, Melbourne, Australia
d
School of Civil Engineering, College of Engineering, University of Tehran, Tehran, Iran
ARTICLE INFO
Keywords:
Intelligent transportation
Customer behavior
Transport policy
Driverless cars
Smart mobility
ABSTRACT
Growing global research utilizes user acceptance models to investigate the public acceptance of automated ve-
hicles (AVs). A growing body of literature suggests it is essential to recognize cultural differences that may in-
uence people's decisions and the intention to use (AVs). While the inuence of perceived safety on AVs adoption
has been examined globally, it has often been overlooked in Australia. To address this knowledge gap, this study
extended the Unied Theory of Acceptance and Use of Technology (UTAUT) model by incorporating perceived
safety and socio-demographic factors in assessing behavioral intention for fully AVs in Australia. This study is the
rst in Australia to include perceived safety in the UTAUT model and look at how different factors like age,
gender, experience, income, education, and travel habits affect people's intention to use technology. The model
was evaluated with Structural Equation Modelling using a dataset of 804 respondents from Australia.
Perceived Safety (PS) holds comparable importance to Social Inuence (SI) and Facilitating Conditions (FC).
Our analysis revealed that younger age groups exhibit a more substantial positive correlation between Perfor-
mance Expectancy (PE) and Behavioral Intention (BI) compared to older age groups. Notably, there are signif-
icant distinctions in the impact of PS on BI between older and younger age groups, as well as between those with
and without prior experience with AVs. Moreover, gender has a moderating effect akin to age in the PE-BI
relationship. Our ndings also reveal that age moderates the relationship between PE and BI, with younger
individuals exhibiting less susceptibility to social inuence compared to older counterparts. Gender also emerges
as a moderator, affecting the relationship between FC and BI. Additionally, income moderates the relationships
between both EE (Effort Expectancy) and FC with BI. However, qualications do not signicantly moderate the
relationships between latent variables and BI.
The multigroup analysis highlights a signicant divergence in the inuence of PE on BI between groups with
no experience and experienced people. Additionally, the study shows that the higher-income group displays a
lower coefcient of FC towards BI, potentially due to their pre-existing knowledge base. The ndings from this
study assist decision-makers by providing insights into public attitudes towards AVs by revealing the key factors
inuencing public acceptance.
1. Introduction
Automated vehicles (AVs) are poised to transform the landscape of
transportation, offering substantial advantages, including heightened
safety, improved accessibility, reduced congestion, and lower emissions
(Chen, Shiwakoti, Stasinopoulos, & Khan, 2022). As a result, the
advancement of AVs has drawn considerable interest from a diverse
array of stakeholders, encompassing technology industry leaders,
automotive manufacturers, and mobility service providers. According to
the analysis conducted by (Chen, Stasinopoulos, Shiwakoti, & Khan,
2023), it is anticipated that approximately 20% of commuters will uti-
lize AVs as a transportation mode by 2048.
As we witness the accelerated integration of AVs in the coming de-
cades, both governmental entities and private enterprises must proac-
tively engage with this disruptive technology, leveraging its potential to
enhance societal well-being. Thus, it becomes imperative to
* Corresponding author at: RMIT University, City Campus, 124 La Trobe Street, Melbourne 3000, Victoria, Australia.
E-mail address: nirajan.shiwakoti@rmit.edu.au (N. Shiwakoti).
Contents lists available at ScienceDirect
Research in Transportation Business & Management
journal homepage: www.elsevier.com/locate/rtbm
https://doi.org/10.1016/j.rtbm.2024.101147
Received 22 November 2023; Received in revised form 22 April 2024; Accepted 28 May 2024
Research in Transportation Business & Management 56 (2024) 101147
2
comprehend individuals' perceptions of autonomous vehicles and the
subsequent ripple effects on societal sentiment. Nevertheless, a con-
spicuous research gap exists regarding the multifaceted factors inu-
encing the adoption of AVs, encompassing demographic considerations,
tangible experiences with AVs, safety awareness, technical prociency,
risk assessment, and driving conditions. This gap is particularly notice-
able within the Australian context. Furthermore, our understanding of
the psychological dimensions governing individuals' sentiments towards
AVs and their potential impact on broader public opinion and behavioral
intentions remains limited.
Previous research, such as the work by (Chen, Khan, Shiwakoti,
Stasinopoulos, & Aghabayk, 2023), utilized the revised Technology
Acceptance Model (TAM) to study the factors impacting behavioral in-
tentions concerning the adoption of AVs in Australia. However, the
study lacks the consideration of perceived safety (PS) and does not
investigate the potential impacts of socio-demographic factors on their
acceptance. While prior Australian research delved into safety concerns
surrounding Automated Vehicles (AVs) through interviews conducted in
Brisbane, yielding positive perceptions of Perceived Safety (PS) via
thematic analysis, there remains a gap in systematic quantitative and
qualitative investigations to comprehensively assess the impacts of PS
(Swain, Truelove, Rakotonirainy, & Kaye, 2023). (Kaye, Lewis, Forward,
& Delhomme, 2020) investigated the impact of socio-demographic
factors on AVs acceptance in Australia; however, it lacked a compre-
hensive quantitative and qualitative analysis to systematically examine
these impacts. The study solely utilized regression methods to explore
the relationships between UTAUT constructs and predictors of in-
tentions separately, focusing primarily on age, gender, and pre-existing
knowledge. Similarly, (Cunningham, Regan, Horberry, Weeratunga, &
Dixit, 2019) investigated the relationship between socio-demographic
factors and selected dependent variables in Australia, yet their anal-
ysis was more segmented and lacked a systematic exploration of how
these factors affect the variables. Moreover, a recent study in Queens-
land, Australia (Li, Kaye, Afghari, & Oviedo-Trespalacios, 2023),
explored AVs acceptance among drivers, pedestrians, and cyclists,
considering the impact of age, gender, and exposure time through
regression analysis. However, this study did not investigate the moder-
ating effects on important variables within their Theory of Planned
Behavior (TPB) model. While (Golbabaei, Yigitcanlar, Paz, & Bunker,
2023) employed logistic regression to examine the inuence of socio-
demographic factors on attitudes towards autonomous shuttle buses in
Southeast Queensland, there is a notable absence of comprehensive
analysis regarding how these factors moderate behavioral intentions
towards AVs. Recent studies in China (Yao et al., 2023), Indonesia
(Prasetio & Nurliyana, 2023) and Spain (Montoro et al., 2019)empha-
sized that perceived safety could play a signicant role in AVs adoption.
While perceived safety has been examined globally, it is essential to
recognize that cultural differences may inuence people's decision
making (Frank, Chrysochou, Mitkidis, & Ariely, 2019; Maxmen, 2018)
and intention to use AVs. More studies in other countries may provide
insights into geographical bias as cultural differences play a critical role
in the adoption of AVs (Yun, Oh, & Myung, 2021). Further, there is a
shortage of studies that consider additional external factors for fully AVs
adoption such as social inuence, facilitating conditions, and perceived
safety, and their relationship with important socio-demographic factors
moderating effects, specically within the Australian context.
Hence, our study aims to integrate perceived safety (PS) into the
UTAUT model and quantitatively and qualitatively analyze socio-
demographic moderating effects, particularly in the Australian
context. To the best of our knowledge, this is the rst study in Australia
that incorporates perceived safety into the UTAUT model and system-
atically tests the soci-demographic factors' moderating impacts,
including age, gender, prior experience, income, qualications and
travel behavior. Recognizing the diversity within Australian society, this
comprehensive examination allows us to uncover potential disparities in
adoption patterns across different demographic groups, thereby
informing targeted strategies for promoting greater acceptance and
uptake of AVs. Acknowledging these socio-demographic effects enables
governments to plan strategically for the future by comprehending
public perceptions and their intended behaviors.
Moreover, our research emphasizes the importance of cultural dif-
ferences in the adoption of AVs and their implications for mitigating
geographic biases. By studying the Australian context, we contribute
valuable insights that can help bridge cultural divides and facilitate a
more universal understanding of the factors driving AV adoption. As AV
technology continues to evolve on a global scale, our ndings serve to
enrich the body of knowledge on cultural inuences and pave the way
for more inclusive approaches to promoting its adoption worldwide.
This paper is organised as follows. In section 2, we provide a
comprehensive review of the current literature, including UTAUT and
socio-demographic factors. Section 3 demonstrates the methodology of
this study consisting of data source, measurements of the latent variables
as well as data analysis method. In section 4, we present the ndings
containing descriptive analysis of the survey, structural equation
modelling of the revised UTAUT model and socio-demographic factors
moderating effects. Based on the results of section 4, section 5 discusses
the results by comparing the results of the revised UTAUT model and
socio-demographic factors moderating effects with the recent studies.
Section 6 presents the conclusion and study's limitations. The study's
acronyms are presented in Table A1.
2. Theoretical background on determinants of AVs adoption
2.1. Extended UTAUT
The Unied Theory of Acceptance and Use of Technology (UTAUT),
alongside its predecessors, the Technology Acceptance Model (TAM)
and the Theory of Planned Behavior (TPB), constitutes a robust frame-
work widely employed in the realm of technology adoption research.
UTAUT model was established to explain the adoption of new technol-
ogy in organisational context (Venkatesh, Morris, Davis, & Davis, 2003),
and the model has been extended to adapt to a certain context and
improve the prediction power (Venkatesh, Thong, & Xu, 2012). A recent
study (Rejali, Aghabayk, Esmaeli, & Shiwakoti, 2023) has also compared
the three approaches (UTAUT, TPB and TAM) in regard to the accep-
tance of AVs in Iran. The study found behavioral intentions towards AVs
were the strongest predictor (Rejali et al., 2023). These models have
been instrumental in shedding light on the intricate dynamics of tech-
nology acceptance and use by individuals.
The UTAUT theoretical framework posits that the use of technology
hinges on behavioral intent. This intent, in turn, is inuenced directly by
four critical factors: performance expectations (PE), effort expectation
(EE), social inuence (SI), and facilitating conditions (FC) (Venkatesh
et al., 2003). In addition, UTAUT is widely used in the transport industry
to investigate the complex relationship between variables and their
impacts towards the intention to adopt new technology (Cai, Yuen, &
Wang, 2023; Jain, Bhaskar, & Jain, 2022; Madigan et al., 2016). Using
this model as a foundation, the subsequent literature review serves as
the theoretical backdrop by using the recent applications.
2.2. Hypothesis development
2.2.1. Behavioral intention
A study conducted in the USA also investigated the impact of mul-
tiple variables on the behavioral intention to utilize fully AVs through
the application of structural equation modelling. The study's ndings
highlighted the signicant importance of instrumental, symbolic, and
affective motives (Benleulmi & Ramdani, 2022). Another study incor-
porated several constructs to understand the role of behavioral intention
to adopt online shopping behavior by extending UTAUT model (Erjavec
& Manfreda, 2022). Likewise, employing the UTAUT model, a research
study examined the behavioral intention of university students towards
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
3
adopting mobile banking (Tan & Leby Lau, 2016).
Assessing the actual usage of new technology is pivotal, but it is
challenging before the technology becomes a reality. Hence, behavioral
intention serves as a key indicator for technology acceptance, given
established connections between intention and actual use. The subse-
quent sections detail the extended UTAUT variables and associated
hypotheses.
2.2.2. Performance expectancy
To gain insights into the receptivity of novel technologies and in-
dividuals' behavioral inclinations towards them, a recent investigation
by (Kilani, Kakeesh, Al-Weshah, & Al-Debei, 2023) underscored the
signicance of performance expectancy (PE) and effort expectancy (EE)
in shaping the intention to adopt e-wallets within the Jordanian context,
employing the UTAUT2 model as the analytical framework. Further-
more, an examination of the behavioral intentions related to the adop-
tion of urban air mobility within the South Korean context is conducted,
as presented in (Lee, Bae, Lee, & Pak, 2023), employing the UTAUT
model. This study places particular emphasis on the role of performance
expectancy (PE) and facilitating conditions (FC) in fostering favorable
attitudes and behavioral inclinations. Consequently, the study proposes
developing targeted interventions to enhance societal acceptance of
urban air mobility. In the realm of communication technology and the
media industry, (Soren & Chakraborty, 2024) employed the UTAUT 2
model to elucidate the determinants inuencing the intention to utilize
over-the-top platforms. Their study highlights the paramount signi-
cance of PE, EE, and habitual usage as the foremost determinants of
behavioral intention in this context. Therefore, this study assumes that:
H1.: Performance expectancy positively inuences behavioral
intention.
2.2.3. Effort expectancy
(Singh, Singh, Singh, & Higueras-castillo, 2023) also utilized UTAUT
2 and the norm activation model by incorporating PE, EE, social inu-
ence (SI), facilitating condition (FC), hedonic motivation, price value
and habit to evaluate the EV's adoption intention. Although there is a
difference between EV and AV, the similar concept can be used as a
reference to identify the variables that affect adoption intention.
Moreover, in Thaliand, a similar study incorporates EE to use structural
equation modelling to investigate behavioral inuences of carpool
adoption for educational trips (Lowe & Piantanakulchai, 2023). There-
fore, this study assumes that:
H2.: Effort expectancy positively inuences behavioral intention.
2.2.4. Social inuence
The study's ndings highlight the signicant role of social inuence
variables, particularly those related to live social interactions and hy-
pothetical adoption rates within social networks, in explaining the
heterogeneity of bike-sharing adoption (Manca, Sivakumar, & Polak,
2022). Social inuence is crucial in understanding transport technolo-
gies, especially in the early stages of marketization. A study found that
initial trust and social inuence played signicant roles in determining
whether people accept automated vehicles (AVs), highlighting the
importance of targeting inuential individuals to promote AV accep-
tance(Zhang et al., 2020). Therefore, H3 is hypothesized in this study:
H3.: Social inuence positively inuences behavioral intention.
2.2.5. Facilitating condition
Facilitating conditions (FC) are vital in understanding transport
technologies, as evidenced by a study focusing on Electronic Vehicle
Diagnostic Technology (EVDT) in Ghana. This study's ndings revealed
that technology facilitating conditions were the most inuential pre-
dictors of mechanics' intention to use EVDT, emphasizing the practical
signicance of such conditions in the adoption of transport-related
innovations(Turkson, Atombo, Akple, & Tibu, 2023). Additionally, FC
plays crucial role in understanding acceptance of AVs as they affect in-
dividuals' intentions to use AVs, with moderating variables like de-
mographic characteristics, psychological traits, and driving behavior
(Naderi & Nassiri, 2023). Hence, in this study, we propose hypothesis
H4:
H4.: Facilitating conditions positively inuences behavioral
intention.
2.2.6. Perceived safety
Perceived safety (PS), including cognitive and emotional di-
mensions, is crucial for fostering widespread adoption of AVs (Prasetio
& Nurliyana, 2023).(Nair & Bhat, 2021) investigated the factors inu-
encing PS regarding sharing roads with AVs in the United States. Their
study concluded that the adoption of AV technology would negatively
impact PS. Survey results revealed that approximately 40% of re-
spondents considered AVs somewhat safe, while 33% expressed reser-
vations, characterizing them as “not too safe.” Another study, based on
student data from a university in China, found that PS signicantly
predicted BI within the TAM framework (Yao et al., 2023). Despite trust
being the primary factor inuencing AV adoption, research from China
indicated that perceived safety risks negatively affected AV acceptance
through trust mediation (Zhang et al., 2019). In Indonesia, PS is vital for
the adoption of AVs and PS not only includes cognitive and emotional
safety for individual people but also considers the safety system of AVs,
especially privacy cybersecurity (Prasetio & Nurliyana, 2023). Hence,
research conducted in Pennsylvania, USA, employing an ordered probit
model, demonstrated that pedestrians and bicyclists with increased
exposure to AVs exhibit heightened levels of PS when sharing roads with
AVs (Rahman, Dey, Pyrialakou, & Das, 2023).
Additionally, PS signicantly inuences drivers' intention to use
them, with factors like demographics, driving habits, and information
technology interaction also playing key roles, as highlighted in (Mon-
toro et al., 2019). Therefore, in this study, we suggest Hypothesis H5:
H5.: Perceived safety positively inuences behavioral intention.
2.3. Socio-demographic moderating effects
Understanding the behavioral intention of AVs involves recognizing
the signicance of socio-demographic moderating effects. These effects
play a pivotal role in shaping individuals' intentions towards AV adop-
tion, as they reect how various demographic factors inuence people's
willingness to embrace this emerging technology. Examining these
moderating inuences provides valuable insights into the nuanced dy-
namics of AV acceptance within diverse populations. As an illustration, a
study conducted by (Foroughi et al., 2023) delved into the moderating
impacts of compatibility, which seeks to assess how seamlessly an
innovation aligns with individuals' existing social and technical sur-
roundings, particularly in the context of AV adoption intention. Addi-
tionally, in the context of Australia, as proposed by (Golbabaei,
Yigitcanlar, Paz, & Bunker, 2022), further investigation into under-
standing public attitudes towards autonomous shuttles should explore
the effects of socio-demographic factors such as age, gender, education,
and income. Hence, we have explored the following socio-demographic
moderating effects to comprehensively assess their quantitative and
qualitative impacts on the behavioral intention towards AVs.
2.3.1. Age
Age signicantly inuences AV adoption in Germany, where accep-
tance is estimated at around 10 years for SAE L3 AVs and 20 years for L5
AVs (Weigl, Eisele, & Riener, 2022). The study highlights varying will-
ingness to pay among different age groups, underlining the importance
for manufacturers and policymakers to consider age-related factors
when promoting AV adoption. Moreover,
age is a signicant factor affecting preferences for different
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
4
congurations of AVs, including adoption, ownership, and shared usage
(Nair, Astroza, Bhat, Khoeini, & Pendyala, 2018). This nding is based
on rank ordered probit modelling applied to data from the 2015 Puget
Sound Regional Travel Study. Furthermore, the research conducted by
(Panagiotopoulos & Dimitrakopoulos, 2018) highlights age as a pivotal
factor in the acceptance of AVs, noting that older individuals are less
inclined to adopt this technology. However, the study did not explore
the underlying reasons for this disparity.
2.3.2. Gender
Another study explores how age and gender inuence AV acceptance
through moderating factors like travel needs, affordability, and expo-
sure to AV knowledge (Wang, Li, Wang, & Wyatt, 2022). The results
indicate that males with greater travel needs, and affordability tend to
accept AVs more, and younger adults are generally more accepting due
to higher exposure to AV technology. Moreover, this study explores
crucial factors inuencing the adoption of autonomous minibuses in
public transport, highlighting that participants' perceptions of safety,
environmental friendliness, and the pleasantness of their rst ride play
signicant roles, with gender moderating some of these effects (Bern-
hard, Oberfeld, Hoffmann, Weismüller, & Hecht, 2020).
2.3.3. Prior experience
Understanding people's readiness to embrace AVs is challenging
because these vehicles are not yet widely accessible for individuals to
gain rsthand experience, making them less familiar compared to
ubiquitous technologies like smartphones or computers (Mack et al.,
2021). In developing countries like Indonesia, AV technology adoption
is hindered by infrastructure gaps and limited awareness. A survey of
629 respondents indicates a preference for level 2 AV technology, with
demographic factors like gender, education, employment, familiarity,
and previous AV experience inuencing acceptance (Nurliyana, Lestari,
Prasetio, & Belgiawan, 2023). While a previous study explored the sig-
nicant connection between experience and shared AV adoption in the
US, it did not elaborate on how experience specically inuences
adoption in this context (Zhang et al., 2020).
2.3.4. Income
Middle-income travellers between 20 and 39 years old are more
likely to adopt Shared AVs, with travel time and walking time being
signicant factors in adoption (Lim, 2021). The research discusses the
adoption of new technology and its relationship with income by exam-
ining the effects of skill-biased technical change in frontier economies
(Schiopu, 2015). It suggests that the level of income (or skill bias) can
impact how countries adopt and benet from new technology. Higher
income or skills may lead to more successful adoption and utilization of
new technology, while lower income or skills may result in lower-quality
adoption and outcomes.
2.3.5. Qualications / education
(Wali & Khattak, 2022) suggest that their ndings could inform
policies to boost AV car adoption. Using a behavioral model, these
policies would cater to the needs of specic groups like older in-
dividuals, the unemployed, and those with less education. Higher
educational attainment is positively associated with pro-technology at-
titudes, favorability towards environmental considerations and collab-
orative consumption, ease of use perception, and perceived commuting
benets of AVs. This suggests that individuals with higher education
levels are more likely to embrace and adopt AV technology (Acheam-
pong & Cugurullo, 2019).
2.3.6. Travel behavior
The Singapore study reveals that attitudes and current travel
behavior signicantly inuence the adoption of autonomous mobility-
on-demand services (AMOD). People with a positive evaluation of
ridehailing and current ridehailing users are more inclined to choose
AMOD, while current car drivers are also more likely to adopt it. In
contrast, users of public transit show lower interest in AMOD adoption
(Mo, Wang, Moody, Shen, & Zhao, 2021). In addition to that, the
analysis also considered various predictor variables, including self-rated
AV travel experience, current travel behavior, experience with advanced
driver assistance systems (ADAS), and socio-demographic factors
(gender and age) (Lehtonen et al., 2022).
Fig. 1 provides an overview of the proposed UTAUT model, including
moderating effects.
3. Methodology
3.1. Data source
Qualtrics, a research rm specializing in online survey administra-
tion, gathered data from 804 Australian participants aged 18 and older.
The sample was structured to mirror the Australian population's age,
gender, and regional distribution, following the Australian Bureau of
Statistics (ABS) 2021 census data.
The survey questionnaire aimed to assess perceptions of AVs and
transport mode preferences in different scenarios. It underwent pilot
testing with input from three academics and two transportation industry
professionals, leading to modications for the nal questionnaire. This
questionnaire comprises four sections: socio-demographic inquiries,
hypothetical transport mode choice scenarios, inquiries about percep-
tions of automated cars, and motivation-related questions regarding
AVs. Before the survey, a brief introduction was presented in the ques-
tionnaire to enhance participants' comprehension of fully AVs and
highlight the distinctions between fully AVs and conventional cars.
Subsequently, data quality was ltered based on criteria like median
response time and inconsistencies in participant responses (e.g., select-
ing "no driving experience" while indicating "private vehicles" as their
primary mode). The “full launch” of the survey occurred from mid-
October to early November 2022, with survey item order randomized
to reduce biases. All participant information remained anonymous and
received approval from the University Human Research Ethics Com-
mittee (Review Reference: 2022–25,201–18,632). On average, partici-
pants spent 16 min on the survey.
The survey included 804 participants from various Australian states,
with the majority from New South Wales (33.3%) and Victoria (26.1%),
followed by Queensland (19.8%). Gender distribution was nearly equal,
with 49.0% males and 50.4% females. The largest age group was 25 to
34 (21.4%), followed by 45 to 54 (18.4%) and 35 to 44 (17.1%). These
demographics are closely aligned with ABS 2021 data.
3.2. Measurements of the latent variables
Table 1 includes the variables and associated measurement items.
Among the six variables, one serves as the outcome variable, gauging the
intention to use AVs. Each variable comprises multiple measurement
items, and respondents provided ratings on a Likert scale ranging from 1
to 5. Here, one signies “strongly disagree,” while ve denotes “strongly
agree.”
3.3. Data analysis
Structural equation modelling (SEM) is used to analyze the structural
relationship of the proposed UTAUT model. Conrmatory factor anal-
ysis (CFA) assessed the suitability of structural equation modelling,
employing Normed Fit Index (NFI), Chi-square value, Root Mean Square
Error Approximation (RMSEA), and Standardized Root Mean Square
Residual (SRMR) as goodness-of-t indices, following recommendations
from (Kline, 2023). As depicted in Table 2, the SRMR and NFI values
obtained were 0.068 and 0.913, respectively, indicating a satisfactory
model t. Additionally, following the recommendations of (Schuberth,
Rademaker, & Henseler, 2023) and (Shi & Maydeu-Olivares, 2020),
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
5
SRMR and NFI are considered the most appropriate indices for assessing
model t in SEM using SmartPLS4. An SRMR value below 0.08 is
generally deemed acceptable, while NFI values range from 0 to 1, with
higher values indicating a better t (Kline, 2023). Convergent validity
ensured signicant factor loadings (>0.7) for each measured item, and
an Average Variance Extracted (AVE) is above 0.5 (Fornell & Larcker,
1981). Discriminant validity examined distinctions between model
factors. Internal consistency relied on Composite Reliability (CR) and
Cronbach's alpha (>0.7). Structural models were evaluated through
Partial Least Squares (PLS) using SmartPLS4 software, applying a
bootstrap procedure with 5000 sub-samples to assess the signicance of
theoretical constructs.
4. Results
This section presents the ndings encompassing descriptive analysis,
measurement model assessment, structural model evaluation, and the
investigation of moderating effects for the six socio-demographic factors.
4.1. Descriptive analysis
The survey's descriptive analysis seeks to unravel fundamental
questions in exploring AV adoption and its socio-demographic impacts.
It delves into the usage of AVs for the rst and last mile of transportation,
aiming to understand the population's reliance on this mode. Addi-
tionally, it investigates prevailing general opinions about AVs and delves
into the role of willingness to pay, a crucial aspect inuencing BI.
Furthermore, this analysis anticipates revealing the preferred
commuting choices for everyday work once AVs enter the market. By
unveiling insights gleaned from the survey's descriptive analysis, this
research aims to uncover the intricate connections between socio-
demographic factors and the acceptance of AVs in section 4.3.
As depicted in Fig. 2, participants were asked about their mode
choice for daily commuting to the train station when they use the train
as their primary mode of transportation. The preeminent choice among
respondents (62.79%) remains the use of a car when parking facilities
are accessible. Following this, walking (23.24%) is the next preferred
Fig. 1. Proposed UTAUT model, including moderating effects.
Table 1
Variables and measurement items.
Variables Measurement items Sources
Performance expectancy
(PE)
1. Using automated cars can make my daily life and work more efcient (PE1)
2. I believe that the use of automated cars can improve trafc quality (PE2)
3. I believe that the use of automated cars can improve trafc safety (PE3)
4. The use of automated cars is benecial to improving both my living and working conditions (PE4)
(Sung, Jeong, Jeong, & Shin, 2015)
Effort expectancy (EE) 1. I will nd automated cars easy to use (EE1)
2. It will be easy for me to become skilful at using automated car (EE2)
3. I will nd it easy to get fully automated car to do what I want it to do (EE3)
4. Learning to use fully automated car will be easy for me (EE4)
(Venkatesh et al., 2012; Ghalandari,
2012)
Social inuence (SI) 1. I will easily be inuenced by others for the opinion of automated cars (SI1)
2. People whose opinions are important to me would like automated cars too (SI2)
3. I think I am more likely to use the automated cars if my friends or family used it (SI3)
(Zhou et al., 2019)
Facilitating condition (FC) 1. I would nd the technical assistance for automated cars to be available from specic individuals or
groups (FC1)
2. I would have the knowledge necessary to use an automated car (FC2)
3. I would know how to supervise an automated car (FC3)
(Zhou et al., 2019)
Perceived safety (PS) 1. I would feel safe inside the automated cars (PS1)
2. I believe that the use of automated cars can improve trafc safety (PS2)
(Madigan et al., 2016)
Behavior intention (BI) 1. I intend to use the automated car (BI1)
2. When I see automated cars on the roads, I would intend to adopt them (BI2)
3. I plan to use automated cars in the future for my workplace trip (BI3)
4. I intend to use automated cars in future if they are available in the market (BI4)
5. I would travel in automated cars with other passengers as well (BI5)
(Choi & Ji, 2015)
(Nees, 2016)
(Venkatesh, 2000)
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
6
mode, particularly when dedicated footpaths are accessible. Bus and
bicycle transportation modes were less favoured, with 9.77% and 4.20%
of respondents opting for them, respectively. This suggests that AVs
could serve as part of a shared transportation system to meet various
people's needs, especially since most individuals still prefer using cars to
reach train stations (Miller, Chng, & Cheah, 2022).
Similar to the ndings presented in Fig. 2, as illustrated in Fig. 3,
survey participants were inquired about their primary mode of trans-
portation for daily commuting purposes. A notable majority of re-
spondents (68.36%) indicated their preference for private car
ownership, while public transport options, encompassing rail, bus, and
tram, collectively constituted 19.65% of the choices. Furthermore,
consistent with the observations made in Fig. 2, lighter modes of
transportation, such as cycling and motorcycle use, were among the
least popular options.
This raises another question regarding the general perception of AVs,
which is framed as “What is your overall opinion of AVs?” While the
majority of the surveyed individuals (37.58%) maintain a neutral stance
on AVs, 26.21% of respondents express a somewhat positive opinion, in
contrast to the 16.44% who hold a somewhat negative opinion.
Hence, there is an increase in the number of individuals expressing a
favorable perspective that aligns with the ndings, which reveals a
corresponding rise in the proportion of respondents willing to allocate
higher percentages of additional payment. Specically, the data exhibits
percentages of 27.81%, 25.71%, and 20.89% for willingness to pay in-
crements of 0 to 10%, 10% to 20%, and 20% or more, respectively.
Furthermore, Fig. 4 illustrates the response distribution to the
question, “In the future, which commuting option do you prefer when
automated cars are widely available?” With over 40% of respondents
favoring their own AVs, while approximately 20% preferred traditional
personal cars, in contrast to other combined modes, which garnered a
total of 40% preference.
4.2. Measurement model assessment and structural equation modelling
Table 3 presents the variables and their respective measurement
items. As discussed in Section 3.2, the revised model is depicted in Fig. 5,
which displays path coefcients and their associated p-values in
parentheses.
Before applying structural equation modelling, we conducted an
assessment of the measurement model using the PLS-SEM algorithm. As
shown in Table 3, all factor loadings exceeded 0.7. Furthermore, the
AVE values for all constructs were above 0.5, and both the composite
reliability (CR) and Cronbach's Alpha (Cr.
α
) for each latent construct
surpassed 0.7.
Table 3 also displays that the square root of AVE is greater than the
correlations with other constructs, indicating robust discriminant val-
idity (Fornell & Larcker, 1981). Additionally, the results conrmed that
each item in the model had higher loadings on its respective parent
construct compared to other constructs, thus establishing strong
discriminant validity. These ndings collectively support the reliability
and validity of the construct measures. To assess the signicance of the
theoretical constructs, a bootstrap procedure involving 5000 sub-
samples was employed.
After conrming the suitability of our measurement model, we pro-
ceeded to test our hypotheses. The structural equation model (SEM) was
employed to examine the relationships between variables, and the results
are presented in Fig. 5. Notably, four out of ve hypothesized relation-
ships were statistically signicant. Hypothesis 1, which posited a positive
impact of PE on Attitude towards BI, was supported (β =0.384, t =9.344,
p<0.01). As anticipated, SI had a signicant positive inuence on BI
(Hypothesis 3; β =0.244, t=9.413, p<0.01). Furthermore, FC exhibited a
notable positive effect on BI (Hypothesis 4; β =0.202, t=7.878, p<0.01).
However, Hypothesis 2, which proposed a relationship between EE and
BI, did not yield statistical signicance (p=0.223) with β =0.032.
4.3. Socio-demographic factors moderating effects
Within the framework of our revised UTAUT model, this study in-
vestigates the moderating inuences of socio-demographic factors,
including age, gender, prior experience, income, education, and travel
behavior, on individuals' behavioral intentions regarding technology
adoption. The variation in adoption intentions of AVs across different
socio-demographic cohorts is highlighted in the study by (Golbabaei,
Yigitcanlar, Paz, & Bunker, 2020). These factors introduce a layer of
complexity in understanding how behavioral intentions vary across
diverse user groups. In the subsequent sections, we delve into the spe-
cic details of how each of these socio-demographic variables interacts
with and shapes behavioral intentions in distinct and meaningful ways.
Figs. A1 to A10 in the appendix show the results from moderating effects
and slope analysis for socio-demographic factors.
4.3.1. Age
To assess the moderating impact of age on the variables within the
revised model, we categorized participants into three age groups: 18 to
44 for young adulthood (52.8%), 45 to 64 for middle adulthood
(32.6%), and 65 and above for older adulthood (14.6%), aligning with
previous research(Lachman, 2001). Our analysis via Smart PLS4
revealed that age moderates the relationship between PE and BI with a
coefcient of −0.052, as well as the link between SI and BI with a co-
efcient of 0.015, as illustrated in Fig. A1.
As shown in Fig. A2 and Fig. A3, our slope analysis unveiled the
moderating effect of age on the relationship between PE on BI and SI on
BI, respectively. When comparing different age groups, we found that
younger individuals (“age at -1”) exhibit a stronger positive effect of PE
on BI compared to older individuals (“age at +1”). However, the situ-
ation is different for the inuence of SI on BI, as indicated in Fig. A3.
Table 2
Model Fit Index.
Model Fit Index Value
SRMR 0.068
NFI 0.913
Subsequently, the revised UTAUT model under-
went qualitative and quantitative testing to
explore the moderating effects of socio-
demographic factors. Multigroup analysis was
conducted upon identifying signicant relation-
ships within specied groups.
Fig. 2. Mode choice to the train station.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
7
Here, younger individuals (“age at −1”) seem to be less affected by so-
cial inuence compared to their older counterparts (“age at +1”).
Multigroup analysis serves as a valuable tool for assessing differences
across groups, assuming that heterogeneity exists among them (Cheah,
Thurasamy, Memon, Chuah, & Ting, 2020). Following our earlier age-
based categorization, we conducted a bootstrap multigroup analysis. The
results obtained through the Welch-Satterthwaite test, indicated a lack of
signicant relationship between the young and middle age groups. How-
ever, there were signicant relationships observed between the middle and
old age groups regarding Performance PE and BI (p=0.045), as well as
between the young and old age groups concerning SI and BI (p=0.023).
Consequently, Table 4 presents the results of structural equation modelling
for the young, middle, and old age groups, respectively.
4.3.2. Gender
To evaluate the moderating effects of gender on the variables in the
revised model, we categorized gender into two groups using dummy
variables: 0 corresponds to female (50.8%), while 1 represents male
(49.2%). Our analysis using Smart PLS4 unveiled that gender has a
moderating effect on the connection between PE and BI with a coef-
cient of 0.095 (P =0.020). It also impacts the link between FC and BI
with a coefcient of −0.102 (P =0.014), as depicted in Fig. A4.
As depicted in Fig. A5, our slope analysis revealed the moderating
impact of gender on the relationship between PE and BI, as well as be-
tween FC and BI. When we compared different gender groups, we
observed that males (coded as “1”) demonstrate a stronger positive ef-
fect of PE on BI in comparison to females (coded as “0”). This suggests
that males tend to have more trust in technology, which contributes to
their higher acceptance of its performance. However, a different pattern
emerged in the case of the inuence of FC on BI, as shown in Fig. A6.
Here, it appears that females (coded as “0”) exhibit a stronger correla-
tion between FC and BI compared to their male counterparts.
Fig. 3. Current primary transport mode.
Fig. 4. Preferred daily work commute options when AVs are in the market.
Table 3
Findings from Conrmatory Factor Analysis: Assessing Internal Consistency,
Reliability, and Convergent Validity of the Measurement Model.
Variable Items Convergent validity Cr.
α
Factor loading CR AVE
Perceived expectancy PE1
PE2
PE3
PE4
0.864
0.845
0.847
0.875
0.882 0.736 0.880
Effort expectancy EE1
EE2
EE3
EE4
0.909
0.900
0.878
0.853
0.913 0.784 0.908
Social inuence SI1
SI2
SI3
0.822
0.882
0.871
0.825 0.738 0.822
Facilitating conditions FC1
FC2
FC3
0.692
0.868
0.886
0.756 0.673 0.749
Perceived safety PS1
PS2
0.929
0.905
0.824 0.842 0.813
Behavioral Intentions BI1
BI2
BI3
BI4
BI5
0.857
0.878
0.881
0.905
0.828
0.920 0.757 0.919
Note: Cr.
α
: Cronbach's Alpha; CR: Composite Reliability; AVE: Average Variance
Extracted. SRMR =0.064.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
8
Likewise, in the moderation analysis, Table 5 presented below in-
dicates signicant disparities between females and males regarding EE
on BI and FC on BI. Furthermore, variations are evident in the rela-
tionship between PS and BI across different gender groups. Females
exhibit a stronger relationship (with a coefcient of 0.266) compared to
males (with a coefcient of 0.155). This discrepancy arises from the fact
that females tend to have greater concerns about PS than males, which
enhances the relative importance of PS in inuencing BI among females.
4.3.3. Prior experience
In the survey, respondents were asked about “have you ever ridden
in automated cars”, with approximately 13.2% responding “yes” and
86.8% responding “no.” Our analysis, conducted using Smart PLS4,
unveiled that experience plays a moderating role in the relationship
between PS and BI, with a coefcient of −0.155 (p=0.004), as illustrated
in Fig. A7. This nding is further supported by Fig. A8 (slope analysis),
indicating that individuals who have not experienced (coded as “0”)
riding in automated vehicles are likely to be more inuenced by their
perception of safety concerning their BI. This suggests a need to enhance
safety awareness among individuals without prior AV experience.
The multigroup analysis in Table 6 reveals two signicant distinc-
tions between the “no experience” and “experience” groups, particularly
in terms of the impact of PE on BI, which stands at 0.355 and 0.543,
respectively, as well as the inuence of PS on BI, with coefcients of
0.205 and 0.022, respectively. This discrepancy may be attributed to
individuals with prior experiences who have a better understanding of
the positive effect of AVs, leading to a stronger correlation with AV
acceptance. Much like the ndings from the slope analysis, a noticeable
difference emerges in the case of PS on BI between these two groups.
Individuals with no prior experience may harbor reservations about the
Fig. 5. Revised UTAUT model.
Table 4
Multigroup analysis for different age groups.
Variables Path Coefcient (P value)
Young Middle Old
PE >-BI 0.401(0.000***) 0.434(0.000***) 0.155(0.199ns )
EE >-BI 0.054(0.169ns)0.019(0.657ns )0.019(0.752ns)
SI >-BI 0.198(0.000***) 0.248(0.000***) 0.382(0.000***)
FC >-BI 0.256(0.000***) 0.182(0.000***) 0.170(0.009***)
PS >-BI 0.115(0.026**) 0.142(0.043**) 0.299(0.019**)
It is evident that older individuals tend not to have a signicant connection
between PE and BI, as there is no substantial relationship. Interestingly, EE
shows a very modest relationship with BI across all age groups. In the older age
group, SI strongly inuences BI, with a one-unit increase in SI resulting in a
0.382 unit increase in BI. This suggests that older individuals are particularly
sensitive to social inuence, making them more likely to be inuenced. As for
FC's impact remains consistent across age groups, with younger individuals
having a stronger connection to BI.
Table 5
Multigroup analysis for different gender groups.
Variables Path Coefcient (P value)
Female Male
PE >-BI 0.279(0.000***) 0.471(0.000***)
EE >-BI 0.041(0.310ns)0.040(0.232ns )
SI >-BI 0.247(0.000***) 0.222(0.000***)
FC >-BI 0.266(0.000***) 0.155(0.000***)
PS >-BI 0.205(0.026**) 0.138(0.023**)
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
9
safety aspects of AVs, resulting in a more pronounced relationship in the
“no experience” group with respect to BI.
4.3.4. Income
Within the survey, 36.9% of respondents reported monthly pocket
money in the range of $0 to $3000 (considered low income), while
51.4% indicated earning $3001 to $7000 (reecting middle income).
Furthermore, 11.7% of respondents reported having an income
exceeding $7000 per month, indicating a high-income bracket. Our
analysis, conducted using Smart PLS4, revealed that income acts as a
moderating factor in the relationship between EE and BI, as well as
between FC and BI. The coefcients for these relationships were found to
be 0.075 (p=0.000) and −0.046 (p=0.030), as depicted in Fig. A9.
As depicted in Fig. A10, individuals with lower income (red line)
exhibit a notably stronger moderating effect between FC and BI
compared to those with higher income (green line). This difference can
be attributed to individuals with lower incomes often requiring more
technical assistance, likely due to lower educational levels.
Following the multigroup bootstrap analysis, a noteworthy distinc-
tion emerged between low-income and high-income groups in the
context of FC's impact on BI. Given the relatively low coefcients be-
tween EE and BI, we will refrain from further elaborating on the
moderating effects of their relationship.
Table 7 presents the coefcients and their corresponding signicance
values for low, middle, and high-income groups. In the high-income
group, we observe a stronger relationship between PE and BI, as well
as EE and BI. However, the relationship appears to be weaker for SI, FC,
and PS on BI. Middle-income group behavior closely resembles that of
the low-income group. Notably, for the high-income group, PE emerges
as the most inuential factor in shaping BI, with a substantial coefcient
of 0.524.
4.3.5. Qualications
Regarding the respondents' educational backgrounds, it is note-
worthy that 46.5% of them hold certicates, diplomas, or vocational
degrees, which we have categorized as the “middle qualications
group.” Surprisingly, this proportion closely matches the 47.2% of re-
spondents who possess undergraduate, master's, or Ph.D. degrees,
designated as the “high qualications group.” On the other hand, a
relatively smaller percentage of 6.2% falls under the “low qualications
group,” which includes those who did not complete high school or have
no formal education.
Following the examination of moderating effects based on quali-
cations, it was observed that no signicant relationships exist in terms of
qualications moderating between latent variables and BI. Conse-
quently, there is no requirement to perform multigroup analysis since
there are no discernible differences in relationships across different
qualication groups.
4.3.6. Travel behavior
In response to the question about their primary commuting mode,
most respondents (68.36%) indicated that they prefer using private
vehicles, while the remaining respondents (31.64%) opt for other
modes, including public transport. Upon conducting moderating effects
analysis, no signicant relationship between latent variables and BI was
found in relation to travel behavior. Consequently, similar to quali-
cations, there is no need to engage in multigroup analysis, as there are
no notable differences in relationships across travel behavior groups.
5. Discussions
The subsequent sections explore the new ndings of Perceived Safety
(PS) and the inuence of several moderating factors, particularly the
main socio-demographic elements that have moderated the acceptance
of AVs within the revised UTAUT model.
Only EE shows no signicant relationship towards BI with a very
weak coefcient. PE shows the strongest relationship towards BI fol-
lowed by SI, FC and PS. This aligns with previous research using UTAUT
and TPB (Kaye et al., 2020) that found only PE was a substantial pre-
dictor of BI for individuals in Australia and Sweden. Notably, the study
included only PE, EE, and FC variables. Moreover, the research identi-
ed a notable association between PS and BI with a coefcient of 0.171.
This nding closely aligns with a study conducted by (Yao et al., 2023)
(with a coefcient of 0.135), even though their study employed the
Technology Acceptance Model (TAM).
In addition, a study carried out in Turkey found that Trust and Safety
exhibited the strongest impact on BI (Korkmaz, Fidanoglu, Ozcelik, &
Okumus, 2022), with a coefcient of 0.364 (compared to 0.171 in our
study), followed by PE with a coefcient of 0.183 (in contrast to 0.384 in
our study). The disparity between these two countries may be attributed
to the distinctions between developing and developed nations. Citizens
in developing countries might display greater concern regarding safety
and trust in new technology due to lower levels of education.
5.1.1. Age
Our analysis indicates that younger age has a more signicant pos-
itive inuence on PE and BI compared to older age groups. This nding
aligns with previous research using the revised Technology Acceptance
Model (TAM)(Chen, Khan, et al., 2023), indicating a stronger relation-
ship between attitudes towards using (ATU) on BI due to the ATU's close
correlation with PE. Moreover, study emphasized that younger drivers
with social support, along with lower perceived driving abilities, are
more inclined to accept AVs (Huang, Hung, Proctor, & Pitts, 2022).
Despite differing variables, both ndings highlight age as a crucial factor
in shaping technology acceptance. Fascinatingly, studies conducted in
China found no substantial connection between age and BI concerning
AVs (Zhang, Tao, et al., 2020). This could be attributed to changing
behaviors, driven by technological advancements in AVs, inuenced by
demographic and cultural shifts, considering the study was conducted
ve years ago.
Regarding SI, the data indicates that younger individuals are less
inuenced by peers compared to older individuals. This could be related to
the perception of younger people being seen as more experienced and
reliable, a conclusion in line with ndings from studies on adolescence
(Knoll, Leung, Foulkes, & Blakemore, 2017). Hence, in line with the
research by (Günthner & Proff, 2021), enhancing trust levels among older
age groups could foster acceptance of AVs. Improved trust has the potential
Table 6
Multigroup analysis for different experience groups.
Variables Path Coefcient (P value)
No experience Experience
PE >-BI 0.355(0.000***) 0.543(0.000***)
EE >-BI 0.029(0.298ns) − 0.011(0.893ns)
SI >-BI 0.246(0.000***) 0.216(0.000***)
FC >-BI 0.203(0.000***) 0.197(0.000***)
PS >-BI 0.205(0.026**) 0.022(0.846**)
Table 7
Multigroup analysis for different income groups.
Variables Path Coefcient (P value)
Low Middle High
PE >-BI 0.357(0.000***) 0.394(0.000***) 0.524(0.000***)
EE >-BI −0.061(0.169ns)0.059(0.133ns )0.137(0.023**)
SI >-BI 0.257(0.000***) 0.252(0.000***) 0.191(0.005***)
FC >-BI 0.272(0.000***) 0.187(0.000***) 0.084(0.117ns )
PS >-BI 0.201(0.004***) 0.143(0.009***) 0.108(0.396ns )
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
10
to positively impact SI, a signicant factor associated with BI in the older
age group within our study. In comparison, a study conducted in Singapore
yielded results similar to this research. For instance, the young group had a
coefcient of 0.249 (0.198 in our study) for SI on BI, while the older group
had a coefcient of 0.449 (0.382 in our study) for SI on BI (Wu, Wang, &
Yuen, 2023). In contrast to the analysis in the Himalayan region in India,
both studies also found signicant differences in social inuence among
various age groups. However, there were varying coefcients for social
inuence on behavioral intention (0.027 for ages 21 to 35 and 0.238 for
ages above 35 compared to 0.198 for ages 18 to 44, 0.248 for ages 45 to 64,
and 0.382 for ages above 65 in our study) (Singh et al., 2023). These dif-
ferences in coefcients might be attributed to different age categorization
groups. Notably, the introduction of the new variable (PS), revealed a
signicant difference between older individuals and those in younger or
middle age groups. This suggests that older individuals exhibit greater
sensitivity to safety concerns related to AVs.
Moreover, another intriguing study used the UTAUT model to
explore how age inuenced ChatGPT acceptance and arrived at a similar
conclusion. It found that older individuals were more inuenced by their
peers, reecting the same trend observed in this study, where older in-
dividuals have a stronger coefcient on social inuence compared to
younger individuals, despite examining different new technologies
(Menon & Shilpa, 2023).
5.1.2. Gender
Gender shows a moderating inuence similar to age in the relationship
between PE and BI. It also moderates the link between FC and BI, with
varied coefcients for PE and FC in relation to BI between female and male
groups. Consequently, this indirectly supports the signicant relationship
between gender and BI, which aligns with (Kaye et al., 2020)’s ndings in
Australia. Likewise, akin to the study by (Goldbach, Sickmann, Pitz, &
Zimasa, 2022), gender becomes a key factor in adopting public AVs when
there is an absence of employees aiding in service operation, identied
through regression analysis. Previous research conducted by (Chen, Shi-
wakoti, Stasinopoulos, Khan, & Aghabayk, 2023) using the same dataset
demonstrated that gender signicantly inuences the general opinion
about AVs. This general opinion strongly correlates with the BI in this study
as well. Compared to the research conducted in Germany (Kapser, Abdel-
rahman, & Bernecker, 2021), both studies demonstrate that males exhibit a
higher PE on BI coefcient (0.471 for males versus 0.279 for females in our
study) and (0.332 for males versus 0.153 for females in the other study). PS
exhibits a noteworthy association with BI in both male (coefcient: 0.138)
and female (coefcient: 0.205) groups. However, the gender difference in
sensitivity to PS is less pronounced compared to age groups, with females
displaying greater sensitivity to PS than males.
However, a study from Singapore shows that gender has no signicant
difference in terms of trust and acceptance of Shared AVs (Wu et al., 2023).
Similarly, a study conducted in Hungary using the TAM model for tourism
purposes indicated no signicant relationship between gender and the
intention to use (J´
aszber´
enyi, Miskolczi, Munk´
acsy, & F¨
oldes, 2022).
5.1.3. Prior experience
The study highlights a notable moderating impact of PS on BI among
experience groups. The “no experience” group shows a substantial co-
efcient of 0.205, indicating a stronger inuence of PS on BI. In contrast,
the experience group exhibits much lesser sensitivity to PS, likely due to
their higher knowledge and familiarity with AV safety aspects. The
multigroup analysis also emphasizes a substantial difference in the in-
uence of PE on BI between these groups. These ndings align with
similar observations in Germany and China, indicating that direct
experience can facilitate people to use AVs by increasing their trust
(Goldbach et al., 2022; Xu et al., 2018).
Furthermore, the study also found a similar conclusion regarding
shared AVs. It indicates a strong correlation between current travel ex-
periences with ride-hailing apps and the adoption of AVs (Zhang, Wang,
et al., 2020).
5.1.4. Income
A signicant relationship between income level and attitude towards
using AVs on BI was found in the study by (Chen, Khan, et al., 2023).
Although the model was based on TAM, it indirectly parallels the nd-
ings of this study, which indicates that income moderates the relation-
ship between FC and BI. Furthermore, an American study highlights that
early adopters of AVs typically possess higher incomes and extensive
knowledge about AVs (Hardman, Berliner, & Tal, 2019). This aligns with
our study's ndings, indicating that the higher-income group shows a
lower coefcient of FC towards BI, potentially due to their pre-existing
knowledge base. Similar to the ndings from the study conducted in
New York (Daziano, Sarrias, & Leard, 2017), our multigroup analysis
suggests that the lower-income group exhibits a slightly higher PS co-
efcient compared to the middle income group, potentially due to their
concerns regarding safety issues. In the high-income group, no signi-
cant relationship is observed between perceived PS and BI. This implies
that, for individuals with higher incomes, safety concerns do not directly
inuence the adoption of AVs.
5.2. Practical Implications
The practical implications of our research study are multifaceted,
offering actionable insights for various stakeholders involved in pro-
moting the adoption of AVs in the Australian context. Our ndings
reveal distinct patterns in the factors inuencing AV acceptance across
different demographic groups, highlighting the need for tailored stra-
tegies to address varying preferences and concerns.
One key implication is the importance of targeted education and
communication efforts. Younger and middle-aged drivers exhibit high
sensitivity to PE, suggesting the necessity of tailored AV knowledge
sessions for these groups. Interactive workshops or online tutorials could
be developed to educate them about the performance benets of AVs,
enhancing their understanding and condence in the technology. First
hand exposure to educational and communicative efforts could effec-
tively alter individuals' perceptions of AVs (Golbabaei et al., 2020).
Conversely, older individuals demonstrate greater sensitivity to-
wards PS. Clear demonstrations of AV safety features and benets
tailored specically for this group are essential. For instance, demon-
stration events could showcase safety features like collision avoidance
systems, alleviating safety concerns and increasing condence in AV
technology among older demographics.
Gender-specic communication strategies are also warranted, as fe-
males exhibit distinct preferences in the aspect of PE towards BI. Man-
ufacturers should ensure accuracy in communicating AV performance,
tailoring messages to resonate with the preferences of female audiences.
Advertising campaigns could highlight features that appeal to women
(Walsh, 2010), such as enhanced comfort and convenience for family
outings or commuting.
Furthermore, individuals lacking prior AV experience are signi-
cantly inuenced by PE towards BI. This underscores the importance of
PE informational sessions for this cohort. Educational seminars or online
resources supplemented by test drives or simulators could provide
detailed information about the performance capabilities of AVs,
addressing common misconceptions and concerns.
Qualications may inuence income levels and PE sensitivity. Poli-
cymakers and industry stakeholders should consider these factors when
designing strategies to promote AV adoption. Subsidized AV programs
or nancial incentives could make AV technology more accessible to
individuals from lower-income backgrounds, thereby increasing overall
adoption rates.
Interestingly, our study found that travel behavior does not signi-
cantly affect variables inuencing AV acceptance. This suggests that
targeting individuals based on travel behavior may not be necessary to
foster AV acceptance. Instead, efforts should focus on addressing key
drivers of acceptance, such as performance expectancy and perceived
safety, across diverse demographic groups.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
11
5.3. Limitations
One limitation of this study is that the survey data was based on de-
scriptions of AVs rather than actual trials, potentially impacting data ac-
curacy by not reecting real-world experiences. Additionally, the revised
UTAUT model was validated solely among a sample of Australian citizens.
It is important to note that its validity in other countries may differ, as
highlighted by (Kaye et al., 2020), indicating varying coefcients across
different countries. To enhance the generalizability of our ndings, an
international survey is imperative to evaluate the revised model's validity
across diverse populations. Conducting such research will contribute to a
more comprehensive understanding and applicability of our results. The
other limitation is that the public's perceptions of AV technology may
evolve over time as the technology itself advances. This evolution in public
opinion may impact the adoption and acceptance of AVs. Given the current
lack of fully AVs for public use in Australia, understanding public per-
ceptions towards them relies largely on self-reported preferences, a method
known to carry inherent limitations. Moreover, such surveys are vulner-
able to social desirability bias, wherein respondents may shape their an-
swers to align with perceived social norms or expectations.
6. Conclusions
The study extended the UTAUT model by integrating perceived
safety (PS) to evaluate the behavioral intention (BI) to use fully AVs in
Australia. Within the revised UTAUT model, the research extensively
examined six crucial socio-demographic factors—age, gender, prior
experience, income, qualications, and travel behavior—both quanti-
tatively and qualitatively.
We collected data through a nationwide online survey involving 804
respondents from Australia between September and November 2022.
Our analysis highlighted that PE had the most substantial impact on BI,
followed equally by SI, FC, and PS. Perceived safety (PS), a vital addition
to the UTAUT model, holds comparable importance to SI and FC.
Notably, there are signicant distinctions in the impact of PS on BI be-
tween older and younger age groups, as well as between those with and
without prior experience with AVs.
Descriptive survey analysis revealed Australians' general optimism
towards AVs, evident in their preference for private cars as the primary
mode of transportation. Moreover, around 40% indicated they would
choose private AVs once available. An extensive examination of socio-
demographic factors demonstrated that age, gender, prior experience,
and income inuenced relationships between certain variables and BI.
Multigroup analysis revealed signicant impacts from age and gender on
PE's coefcient on BI. Notably, qualications and primary travel
behavior did not moderate any variables in the revised UTAUT model.
The study shows that age, gender, prior experience and income could
signicantly affect one or more constructs, thus affecting behavioral
intention, especially in the Australian context. Moreover, exploring
various sociodemographic factors at granular levels can offer profound
insights, such as categorizing age and income into three distinct groups.
For instance, a notable disparity appeared between individuals with low
and high incomes regarding the impact of FC on BI. Those with lower
incomes demonstrated a signicantly stronger moderating effect be-
tween FC and BI compared to their higher-income counterparts.
Regarding age, signicant associations were identied between middle-
aged and elderly groups concerning PE and BI, as well as between
younger and elderly groups regarding SI and BI.
Building on the insights garnered from our research, future studies
could delve deeper into several avenues to further enhance our under-
standing of AV adoption dynamics. One potential direction could involve
conducting longitudinal research to track the evolution of public percep-
tions towards AVs over time, particularly in response to advancements in
AV technology and changes in societal attitudes. Additionally, exploring
the intersectionality of demographic factors, such as age, gender, educa-
tion, and income, with cultural inuences on AV acceptance could provide
valuable insights into the nuanced preferences and concerns of different
population segments. Furthermore, investigating the impact of regulatory
frameworks and public policy interventions on AV adoption rates could
shed light on the role of governance in shaping the trajectory of AV inte-
gration. Integrating qualitative methods, such as focus groups or in-
terviews, supplementing with test drives of AVs or through simulators and
quantitative analysis could offer a more comprehensive understanding of
the underlying motivations and barriers inuencing AV adoption de-
cisions. Additionally, investigations following the implementation of
suggested recommendations, such as offering comprehensive safety in-
formation about AVs, could be conducted to assess any resultant changes in
comparison to pre-implementation data, thereby evaluating these pro-
posed interventions.
Another potential area for future research involves exploring the
socio-demographic inuences on shared AVs, considering their signi-
cance in addressing the rst and last-mile problem for AVs. Additionally,
expanding the revised UTAUT model to encompass other variables such
as price value, perceived risk, and attitudes towards digitalization could
be valuable. This extended model would allow for the examination of
socio-demographic moderating effects on these additional variables.
Comparing the results would provide a more comprehensive and
insightful analysis. By addressing these research gaps, future studies can
contribute to the development of tailored strategies and policies aimed
at fostering widespread acceptance and successful integration of AV
technology into society.
CRediT authorship contribution statement
Yilun Chen: Writing – review & editing, Writing – original draft,
Software, Methodology, Investigation, Formal analysis, Data curation,
Conceptualization. Shah Khalid Khan: Writing – review & editing,
Formal analysis. Nirajan Shiwakoti: Writing – review & editing, Su-
pervision, Investigation, Conceptualization. Peter Stasinopoulos:
Writing – review & editing, Supervision. Kayvan Aghabayk: Writing –
review & editing, Investigation.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Table A1
Acronyms.
Abbreviation Explanation
AVs Automated Vehicles
TAM Technology Acceptance Model
UTAUT Unied Theory of Acceptance and Use of Technology
TPB Theory of Planned Behavior
PE Performance Expectation
BI Behavioral Intention
EE Effort Expectation
SI Social Inuence
FC Facilitating Conditions
EVDT Electronic Vehicle Diagnostic Technology
PS Perceived Safety
SEM Structural Equation Modelling
AMOD Autonomous Mobility-on-Demand Services
ADAS Advanced Driver Assistance Systems
ATU Attitude Towards Using
ABS Australian Bureau of Statistics
ABS Australian Bureau of Statistics
CR Composite Reliability
SRMR Standardized Root Mean Square Residual
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
12
Fig. A1. Age moderating effects on the revised UTAUT model.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
13
Fig. A2. Slope analysis of age moderating effects on PE towards BI.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
14
Fig. A3. Slope analysis of age moderating effects on SI towards BI.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
15
Fig. A4. Gender moderating effects on the revised UTAUT model.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
16
Fig. A5. Slope analysis of gender moderating effects on PE towards BI.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
17
Fig. A6. Slope analysis of gender moderating effects on FC towards BI.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
18
Fig. A7. Prior experience moderating effects on the revised UTAUT model.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
19
Fig. A8. Slope analysis of prior experience moderating effects on PS towards BI.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
20
Fig. A9. Income moderating effects on the revised UTAUT model.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
21
References
Acheampong, R. A., & Cugurullo, F. (2019). Capturing the behavioural determinants
behind the adoption of autonomous vehicles: Conceptual frameworks and
measurement models to predict public transport, sharing and ownership trends of
self-driving cars. Transportation Research Part F: Trafc Psychology and Behaviour, 62,
349–375.
Benleulmi, A. Z., & Ramdani, B. (2022). Behavioural intention to use fully autonomous
vehicles: Instrumental, symbolic, and affective motives. Transportation Research Part
F: Trafc Psychology and Behaviour, 86, 226–237.
Bernhard, C., Oberfeld, D., Hoffmann, C., Weismüller, D., & Hecht, H. (2020). User
acceptance of automated public transport: Valence of an autonomous minibus
experience. Transportation Research Part F: Trafc Psychology and Behaviour, 70,
109–123.
Cai, L., Yuen, K. F., & Wang, X. (2023). Explore public acceptance of autonomous buses:
An integrated model of UTAUT, TTF and trust. Travel Behaviour and Society, 31,
120–130.
Cheah, J.-H., Thurasamy, R., Memon, M. A., Chuah, F., & Ting, H. (2020). Multigroup
analysis using SmartPLS: Step-by-step guidelines for business research. Asian. Journal
of Business Research, 10, I–XIX.
Chen, Y., Khan, S. K., Shiwakoti, N., Stasinopoulos, P., & Aghabayk, K. (2023). Analysis
of Australian public acceptance of fully automated vehicles by extending technology
acceptance model. Case Studies on Transport Policy, 14, Article 101072.
Chen, Y., Shiwakoti, N., Stasinopoulos, P., & Khan, S. K. (2022). State-of-the-art of factors
affecting the adoption of automated vehicles. Sustainability, 14, 6697.
Chen, Y., Shiwakoti, N., Stasinopoulos, P., Khan, S. K., & Aghabayk, K. (2023). Exploring
the association between socio-demographic factors and public acceptance towards
fully automated vehicles: Insights from a survey in Australia. IET Intelligent Transport
Systems, 1–19.
Chen, Y., Stasinopoulos, P., Shiwakoti, N., & Khan, S. K. (2023). Using system dynamics
approach to explore the mode shift between automated vehicles, conventional
vehicles, and public transport in Melbourne. Australia. Sensors, 23, 7388.
Choi, J. K., & Ji, Y. G. (2015). Investigating the importance of trust on adopting an
autonomous vehicle. International Journal of Human-Computer Interaction, 31,
692–702.
Cunningham, M. L., Regan, M. A., Horberry, T., Weeratunga, K., & Dixit, V. (2019).
Public opinion about automated vehicles in Australia: Results from a large-scale
national survey. Transportation Research Part A: Policy and Practice, 129, 1–18.
Daziano, R. A., Sarrias, M., & Leard, B. (2017). Are consumers willing to pay to let cars
drive for them? Analyzing response to autonomous vehicles. Transportation Research
Part C: Emerging Technologies, 78, 150–164.
Erjavec, J., & Manfreda, A. (2022). Online shopping adoption during COVID-19 and
social isolation: Extending the UTAUT model with herd behavior. Journal of Retailing
and Consumer Services, 65, Article 102867.
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with
unobservable variables and measurement error. Journal of Marketing Research, 18,
39–50.
Foroughi, B., Nhan, P. V., Iranmanesh, M., Ghobakhloo, M., Nilashi, M., &
Yadegaridehkordi, E. (2023). Determinants of intention to use autonomous vehicles:
Findings from PLS-SEM and ANFIS. Journal of Retailing and Consumer Services, 70,
Article 103158.
Frank, D. A., Chrysochou, P., Mitkidis, P., & Ariely, D. (2019). Human decision-making
biases in the moral dilemmas of autonomous vehicles. Scientic Reports, 9(1), 13080.
Ghalandari, K. (2012). The effect of performance expectancy, effort expectancy, social
inuence and facilitating conditions on acceptance of e-banking services in Iran: The
moderating role of age and gender. Middle-East Journal of Scientic Research, 12,
801–807.
Golbabaei, F., Yigitcanlar, T., Paz, A., & Bunker, J. (2020). Individual predictors of
autonomous vehicle public acceptance and intention to use: A systematic review of
the literature. Journal of Open Innovation: Technology, Market, and Complexity, 6(4),
106.
Golbabaei, F., Yigitcanlar, T., Paz, A., & Bunker, J. (2022). Understanding autonomous
shuttle adoption intention: Predictive power of pre-trial perceptions and attitudes.
Sensors, 22(23), 9193.
Golbabaei, F., Yigitcanlar, T., Paz, A., & Bunker, J. (2023). Perceived opportunities and
challenges of autonomous demand-responsive transit use: What are the socio-
demographic predictors? Sustainability, 15(15), 11839.
Goldbach, C., Sickmann, J., Pitz, T., & Zimasa, T. (2022). Towards autonomous public
transportation: Attitudes and intentions of the local population. Transportation
Research Interdisciplinary Perspectives, 13, Article 100504.
Günthner, T., & Proff, H. (2021). On the way to autonomous driving: How age inuences
the acceptance of driver assistance systems. Transportation Research Part F: Trafc
Psychology and Behaviour, 81, 586–607.
Hardman, S., Berliner, R., & Tal, G. (2019). Who will be the early adopters of automated
vehicles? Insights from a survey of electric vehicle owners in the United States.
Transportation Research Part D: Transport and Environment, 71, 248–264.
Huang, G., Hung, Y.-H., Proctor, R. W., & Pitts, B. J. (2022). Age is more than just a
number: The relationship among age, non-chronological age factors, self-perceived
Fig. A10. Slope analysis of income moderating effects on FC towards BI.
Y. Chen et al.
Research in Transportation Business & Management 56 (2024) 101147
22
driving abilities, and autonomous vehicle acceptance. Accident Analysis & Prevention,
178, Article 106850.
Jain, N. K., Bhaskar, K., & Jain, S. (2022). What drives adoption intention of electric
vehicles in India? An integrated UTAUT model with environmental concerns,
perceived risk and government support. Research in Transportation Business &
Management, 42, Article 100730.
J´
aszber´
enyi, M., Miskolczi, M., Munk´
acsy, A., & F¨
oldes, D. (2022). What drives tourists to
adopt self-driving cars? Transportation Research Part F: Trafc Psychology and
Behaviour, 89, 407–422.
Kapser, S., Abdelrahman, M., & Bernecker, T. (2021). Autonomous delivery vehicles to
ght the spread of Covid-19–how do men and women differ in their acceptance?
Transportation Research Part A: Policy and Practice, 148, 183–198.
Kaye, S.-A., Lewis, I., Forward, S., & Delhomme, P. (2020). A priori acceptance of highly
automated cars in Australia, France, and Sweden: A theoretically-informed
investigation guided by the TPB and UTAUT. Accident Analysis & Prevention, 137,
Article 105441.
Kilani, A. A.-H. Z., Kakeesh, D. F., Al-Weshah, G. A., & Al-Debei, M. M. (2023). Consumer
post-adoption of e-wallet: An extended UTAUT2 perspective with trust. Journal of
Open Innovation: Technology, Market, and Complexity, 9, Article 100113.
Kline, R. B. (2023). Principles and practice of structural equation modeling. Guilford
publications.
Knoll, L. J., Leung, J. T., Foulkes, L., & Blakemore, S.-J. (2017). Age-related differences in
social inuence on risk perception depend on the direction of inuence. Journal of
Adolescence, 60, 53–63.
Korkmaz, H., Fidanoglu, A., Ozcelik, S., & Okumus, A. (2022). User acceptance of
autonomous public transport systems: Extended UTAUT2 model. Journal of Public
Transportation, 24, Article 100013.
Lachman, M. (2001). Adult development, psychology of.
Lee, C., Bae, B., Lee, Y. L., & Pak, T.-Y. (2023). Societal acceptance of urban air mobility
based on the technology adoption framework. Technological Forecasting and Social
Change, 196, Article 122807.
Lehtonen, E., Malin, F., Louw, T., Lee, Y. M., Itkonen, T., & Innamaa, S. (2022). Why
would people want to travel more with automated cars? Transportation Research Part
F: Trafc Psychology and Behaviour, 89, 143–154.
Li, X., Kaye, S.-A., Afghari, A. P., & Oviedo-Trespalacios, O. (2023). Sharing roads with
automated vehicles: A questionnaire investigation from drivers’, cyclists’ and
pedestrians’ perspectives. Accident Analysis & Prevention, 188, Article 107093.
Lim, T. S. (2021). A study of the scheduling effect on shared autonomous vehicles
adoption. Transportation Research Interdisciplinary Perspectives, 10, Article 100394.
Lowe, W. U. A., & Piantanakulchai, M. (2023). Investigation of behavioral inuences of
carpool adoption for educational trips–A case study of Thammasat University.
Thailand. Case Studies on Transport Policy, 12, Article 100970.
Mack, E. A., Miller, S. R., Chang, C.-H., van Fossen, J. A., Cotten, S. R., Savolainen, P. T.,
& Mann, J. (2021). The politics of new driving technologies: Political ideology and
autonomous vehicle adoption. Telematics and Informatics, 61, Article 101604.
Madigan, R., Louw, T., Dziennus, M., Graindorge, T., Ortega, E., Graindorge, M., &
Merat, N. (2016). Acceptance of automated road transport systems (ARTS): An
adaptation of the UTAUT model. Transportation Research Procedia, 14, 2217–2226.
Manca, F., Sivakumar, A., & Polak, J. W. (2022). Capturing the effect of multiple social
inuence sources on the adoption of new transport technologies and services.
Journal of choice modelling, 42, Article 100344.
Maxmen, A. (2018). Self-driving car dilemmas reveal that moral choices are not
universal. Nature, 562(7728), 469.
Menon, D., & Shilpa, K. (2023). “Chatting with ChatGPT”: Analyzing the factors
inuencing users' intention to use the open AI's ChatGPT using the UTAUT model.
Heliyon, 9.
Miller, K., Chng, S., & Cheah, L. (2022). Understanding acceptance of shared autonomous
vehicles among people with different mobility and communication needs. Travel
Behaviour and Society, 29, 200–210.
Mo, B., Wang, Q. Y., Moody, J., Shen, Y., & Zhao, J. (2021). Impacts of subjective
evaluations and inertia from existing travel modes on adoption of autonomous
mobility-on-demand. Transportation Research Part C: Emerging Technologies, 130,
Article 103281.
Montoro, L., Useche, S. A., Alonso, F., Lijarcio, I., Bos´
o-seguí, P., & Martí-belda, A.
(2019). Perceived safety and attributed value as predictors of the intention to use
autonomous vehicles: A national study with Spanish drivers. Safety Science, 120,
865–876.
Naderi, H., & Nassiri, H. (2023). How will Iranian behave in accepting autonomous
vehicles? Studying moderating effect on autonomous vehicle acceptance model
(AVAM). IATSS Research, 47, 433–446.
Nair, G. S., Astroza, S., Bhat, C. R., Khoeini, S., & Pendyala, R. M. (2018). An application
of a rank ordered probit modeling approach to understanding level of interest in
autonomous vehicles. Transportation, 45, 1623–1637.
Nair, G. S., & Bhat, C. R. (2021). Sharing the road with autonomous vehicles: Perceived
safety and regulatory preferences. Transportation Research Part C Emerging
Technologies, 122, Article 102885.
Nees, M. A. (2016). Acceptance of self-driving cars: An examination of idealized versus
realistic portrayals with a self- driving Car acceptance scale. Proceedings of the Human
Factors and Ergonomics Society Annual Meeting, 60, 1449–1453.
Nurliyana, C., Lestari, Y. D., Prasetio, E. A., & Belgiawan, P. F. (2023). Exploring drivers’
interest in different levels of autonomous vehicles: Insights from Java Island.
Indonesia. Transportation research interdisciplinary perspectives, 19, Article 100820.
Panagiotopoulos, I., & Dimitrakopoulos, G. (2018). An empirical investigation on
consumers' intentions towards autonomous driving. Transportation Research Part C
Emerging Technologies, 95, 773–784.
Prasetio, E. A., & Nurliyana, C. (2023). Evaluating perceived safety of autonomous
vehicle: The inuence of privacy and cybersecurity to cognitive and emotional
safety. IATSS Research, 47, 160–170.
Rahman, M. T., Dey, K., Pyrialakou, V. D., & Das, S. (2023). Factors inuencing safety
perceptions of sharing roadways with autonomous vehicles among vulnerable
roadway users. Journal of Safety Research, 85, 266–277.
Rejali, S., Aghabayk, K., Esmaeli, S., & Shiwakoti, N. (2023). Comparison of technology
acceptance model, theory of planned behavior, and unied theory of acceptance and
use of technology to assess a priori acceptance of fully automated vehicles.
Transportation Research Part A: Policy and Practice, 168, Article 103565.
Schiopu, I. (2015). Technology adoption, human capital formation and income
differences. Journal of Macroeconomics, 45, 318–335.
Schuberth, F., Rademaker, M. E., & Henseler, J. (2023). Assessing the overall t of
composite models estimated by partial least squares path modeling. European Journal
of Marketing, 57(6), 1678–1702.
Shi, D., & Maydeu-Olivares, A. (2020). The effect of estimation methods on SEM t
indices. Educational and Psychological Measurement, 80(3), 421–445.
Singh, H., Singh, V., Singh, T., & Higueras-castillo, E. (2023). Electric vehicle adoption
intention in the Himalayan region using UTAUT2–NAM model. Case Studies on
Transport Policy, 11, Article 100946.
Soren, A. A., & Chakraborty, S. (2024). Adoption, satisfaction, trust, and commitment of
over-the-top platforms: An integrated approach. Journal of Retailing and Consumer
Services, 76, Article 103574.
Sung, H.-N., Jeong, D.-Y., Jeong, Y.-S., & Shin, J.-I. (2015). The relationship among self-
efcacy, social inuence, performance expectancy, effort expectancy, and behavioral
intention in mobile learning service. International Journal of u-and e-Service, Science
and Technology, 8, 197–206.
Swain, R., Truelove, V., Rakotonirainy, A., & Kaye, S. A. (2023). A comparison of the
views of experts and the public on automated vehicles technologies and societal
implications. Technology in Society, 74, Article 102288.
Tan, E., & Leby Lau, J. (2016). Behavioural intention to adopt mobile banking among the
millennial generation. Young Consumers, 17, 18–31.
Turkson, R. F., Atombo, C., Akple, M. S., & Tibu, H. M. (2023). Modelling the adoption of
electronic vehicle diagnostic technology for vehicle repairs: A structural equation
modelling approach. Technological Forecasting and Social Change, 191, Article
122449.
Venkatesh, V. (2000). Determinants of perceived ease of use: Integrating control,
intrinsic motivation, and emotion into the technology acceptance model. Information
Systems Research, 11, 342–365.
Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User acceptance of
information technology: Toward a unied view. MIS Quarterly, 425–478.
Venkatesh, V., Thong, J. Y., & Xu, X. (2012). Consumer acceptance and use of
information technology: Extending the unied theory of acceptance and use of
technology. MIS Quarterly, 157–178.
Wali, B., & Khattak, A. J. (2022). A joint behavioral choice model for adoption of
automated vehicle ride sourcing and carsharing technologies: Role of built
environment & sustainable travel behaviors. Transportation Research Part C: Emerging
Technologies, 136, Article 103557.
Walsh, M. (2010). Gender on the road in the United States: By motor car or motor coach?
The Journal of Transport History, 31(2), 210–230.
Wang, S., Li, Z., Wang, Y., & Wyatt, D. A. (2022). How do age and gender inuence the
acceptance of automated vehicles?–revealing the hidden mediating effects from the
built environment and personal factors. Transportation Research Part A: Policy and
Practice, 165, 376–394.
Weigl, K., Eisele, D., & Riener, A. (2022). Estimated years until the acceptance and
adoption of automated vehicles and the willingness to pay for them in Germany:
Focus on age and gender. International Journal of Transportation Science and
Technology, 11, 216–228.
Wu, M., Wang, N., & Yuen, K. F. (2023). Can autonomy level and anthropomorphic
characteristics affect public acceptance and trust towards shared autonomous
vehicles? Technological Forecasting and Social Change, 189, Article 122384.
Xu, Z., Zhang, K., Min, H., Wang, Z., Zhao, X., & Liu, P. (2018). What drives people to
accept automated vehicles? Findings from a eld experiment. Transportation
Research Part C Emerging Technologies, 95, 320–334.
Yao, S., Xie, L., Chen, Y., Zhang, Y., Chen, Y., & Gao, M. (2023). Inuence of perceived
safety in the technology acceptance model. Transportation Research Part F: Trafc
Psychology and Behaviour, 99, 36–51.
Yun, Y., Oh, H., & Myung, R. (2021). Statistical modeling of cultural differences in
adopting autonomous vehicles. Applied Sciences, 11(19), 9030.
Zhang, T., Tao, D., Qu, X., Zhang, X., Lin, R., & Zhang, W. (2019). The roles of initial trust
and perceived risk in public's acceptance of automated vehicles. Transportation
Research Part C Emerging Technologies, 98, 207–220.
Zhang, T., Tao, D., Qu, X., Zhang, X., Zeng, J., Zhu, H., & Zhu, H. (2020). Automated
vehicle acceptance in China: Social inuence and initial trust are key determinants.
Transportation Research Part C Emerging Technologies, 112, 220–233.
Zhang, W., Wang, K., Wang, S., Jiang, Z., Mondschein, A., & Noland, R. B. (2020).
Synthesizing neighborhood preferences for automated vehicles. Transportation
Research Part C: Emerging Technologies, 120, Article 102774.
Zhou, L. L., Owusu-Marfo, J., Asante Antwi, H., Antwi, M. O., Kachie, A. D. T., & Ampon-
wireko, S. (2019). Assessment of the social inuence and facilitating conditions that
support nurses’ adoption of hospital electronic information management systems
(HEIMS) in Ghana using the unied theory of acceptance and use of technology
(UTAUT) model. BMC Medical Informatics and Decision Making, 19, 1–9.
Y. Chen et al.
