The acceptance of conditionally automated cars from the perspective of different road user groups

Abstract

While public acceptance and the acceptance of potential users have already been intensively researched, this study investigates the acceptance of CACs from the point of view of different road user groups, such as pedestrians, cyclists and riders of powered two-wheelers as so-called vulnerable road users (VRUs), as well as the drivers of conventional cars. The study explores a priori road user acceptance of CACs with a specifically developed measuerement tool using an international population survey that was conducted within the framework of the EU Horizon 2020 funded project ‘BRidging gaps for the adoption of Automated VEhicles’ (BRAVE) in the participating EU-countries France, Germany, Slovenia, Spain and Sweden as well as in Australia and the USA, including 5,827 respondents.

Full text

Issue 21(4), 2021, pp. 81-103
https://doi.org/10.18757/ejtir.2021.21.4.5466
EJTIR
ISSN: 1567-7141
http://ejtir.tudelft.nl/
The acceptance of conditionally automated cars from the
perspective of different road user groups
Bernhard Schrauth1
Institute for Empirical Sociology at the Friedrich-Alexander-University of Erlangen-Nuremberg,
Germany
Walter Funk2
Institute for Empirical Sociology at the Friedrich-Alexander-University of Erlangen-Nuremberg,
Germany
Sarah Maier3
Institute for Empirical Sociology at the Friedrich-Alexander-University of Erlangen-Nuremberg,
Germany
Clemens Kraetsch4
Institute for Empirical Sociology at the Friedrich-Alexander-University of Erlangen-Nuremberg,
Germany
The foreseeable advent of conditionally automated cars (CACs)
at SAE Level 3 opens a range of opportunities along with
numerous questions that must be addressed to safely adopt this
new vehicle technology. While public acceptance and the
acceptance of potential users have already been intensively
researched, this study investigates the acceptance of CACs from
the point of view of different road user groups, such as
pedestrians, cyclists and riders of powered two-wheelers as so-
called vulnerable road users (VRUs), as well as the drivers of
conventional cars. The study measures a priori road user
acceptance of CACs using an international population survey
that was conducted within the framework of the EU Horizon
2020 funded project ‘BRidging gaps for the adoption of
Publishing history
Submitted: 16 December 2020
Accepted: 30 September 2021
Published: 16 November 2021
Cite as
Schrauth, B., Funk, W., Maier,
S., & Kraetsch, C. (2021). The
acceptance of conditionally
automated cars from the
perspective of different road
user groups. European Journal
of Transport and
Infrastructure Research, 21(4),
81-103.
1
A: Marienstrasse 2, 90402 Nuremberg, Germany. T: +49 (0) 911 23565 37.
E: bernhard.schrauth@ifes.uni-erlangen.de
2
A: Marienstrasse 2, 90402 Nuremberg, Germany. E: walter.h.funk@ifes.uni-erlangen.de
3
A: Marienstrasse 2, 90402 Nuremberg, Germany. E: sarah.maier@ifes.uni-erlangen.de
4
A: Marienstrasse 2, 90402 Nuremberg, Germany. E: clemens.kraetsch@ifes.uni-erlangen.de

EJTIR 21(4), 2021, pp.81-103 82
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
Automated VEhicles’ (BRAVE) in the participating EU-countries
France, Germany, Slovenia, Spain and Sweden as well as in
Australia and the USA. Including 5,827 respondents, the study
findings disclose a rather positive acceptance of CACs from the
perspective of different road user groups. However, concerns are
also apparent. Results from multivariate analyses indicate that
the acceptance of CACs differs between road user groups in that
VRUs demonstrated lower acceptance than non-automated car
drivers. The role of trust in the new vehicle technology also
appears to be significant. Consequently, future developments of
CACs should also focus on the communication between
automated cars and bystanders (e.g. via external human-machine
interfaces) to reduce uncertainties and promote trust.
© 2021 Bernhard Schrauth,
Walter Funk, Sarah Maier,
Clemens Kraetsch
This work is licensed under a
Creative Commons
Attribution 4.0 International
License (CC BY 4.0)
Keywords: conditionally automated cars, population survey, road user
group-specific acceptance, SAE Level 3, trust, vehicle automation
technology.
1. Introduction
1.1 Technological development and the acceptance of automated vehicle technology
The technological development of automated vehicles is in full swing. Automated vehicles from
SAE Level 3 to SAE Level 5, as defined by SAE International, are already in pilot stages or being
tested. One distinctive feature of this development is that automation from SAE Level 3 onwards
exceeds the limit that drivers must always drive during the journey: In certain driving situations,
the machine is the executing and responsible system, and the human driver is the fallback system
(see SAE International, 2018). Conditionally automated cars (CACs) corresponding with SAE
Level 3 will be the first step towards automated driving, representing an upcoming paradigm shift
in which the driver hands over the driving task to the car. Though in SAE Level 3 technology, the
CAC being in automated mode requires the driver to be vigilant while attending to other tasks
such as reading or using a smartphone etc. The driver must be able to take over the wheel and the
pedals at any time. By definition, this restriction is only lifted in the later SAE levels 4 or 5.
According to experts, the market-wide introduction of cars with SAE Level 3 technology may be
expected within a few years, albeit at first in special traffic situations such as on motorways
(Eriksson, 2021).
Despite the rapid progress in the last decade, the development of market-ready CACs still faces
technological challenges, such as the vigilance on the part of the driver and the critical situation
that follows take-over requests from the automated system (Banks et al., 2018; Gold, Happee &
Bengler, 2018). Other problems refer to the interaction with other road users that remain to be
defined (Straub & Schaefer, 2019). While the introduction of automated driving opens a range of
societal opportunities, this technological challenge also creates a multitude of political and social
problem areas (Fagnant & Kockelman, 2015; Johnsen et al., 2018; Milakis, van Arem & van Wee,
2017). A successful introduction of CACs, however, requires acceptance by potential users or
buyers as well as by society and other road users. Concerns that are not taken seriously or
insufficient clarification of regulatory issues could lead to a negative attitude towards CACs,
potentially preventing their widespread introduction. Moreover, and along the same lines,
researchers consider the widespread acceptance of CACs in society an important prerequisite, as
the benefits of automated vehicle technology will only materialise if this technology is widely
disseminated (Cunningham & Regan, 2020).

EJTIR 21(4), 2021, pp.81-103 83
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
While public acceptance and potential users’ approval of the technology have already been
intensively researched, the acceptance of automated driving vehicles from the point of view of
different road user groups, including vulnerable road users (VRUs) such as cyclists and
pedestrians, has received scarce attention to date (Deb et al., 2017; Hulse, Xie & Galea, 2018; Saleh,
Hossny & Nahavandi, 2017). Pedestrians, cyclists or riders of powered two-wheelers (PTW riders)
have most often been included in the discussion when their communication or interaction with
automated vehicles have been the object of research (Merat et al., 2018; Stanciu et al., 2018).
However, the research and development of automated vehicle technology might consider the
drivers of non-automated vehicles, cyclists and pedestrians as ‘bystanders’ (Scholtz, 2003) whose
acceptance must also be comprehensively taken into account in the event of this impending change
in road traffic. In this sense, the technological development of automated vehicles must follow the
principle of a Pareto-optimal improvement (Diekmann, 2016). That is, it must be of general interest
that the introduction of CACs must not be limited only at increasing the road safety and comfort
of the ‘drivers’ of automated vehicles. At the same time, other road users like pedestrians, cyclists,
PTW riders and drivers of non-automated cars must not be placed in a worse position and exposed
to any new safety risks or other disadvantages created upon the introduction of automated vehicle
technology.
1.2 Study aims
Against that background, the first aim of this study is to investigate the acceptance of CACs from
the perspective of the different road user groups. For this purpose, this paper proposes a
measurement tool for road user acceptance that captures the a priori acceptance of CACs from the
perspective of other road users and explores the mentioned research gap. Subsequently, as a
secondary goal, the research examines whether road user groups differ in their a priori acceptance
of automatic vehicles with SAE Level 3 technology. In this regard, a special focus will be placed on
the acceptance of CACs on the part of VRUs and whether VRUs differ from each other as well as
from drivers of manually driven cars.
The study bases its findings on a population survey from the research project ‘BRidging gaps for
the adoption of Automated VEhicles’ (BRAVE), funded by the European Union’s Horizon 2020
research and innovation programme (No. 723021). This multidisciplinary research project,
including partners from seven countries, evaluated the needs and concerns of all road users
affected by the introduction of CACs and sought to encourage technological improvements
accordingly.
In this paper, chapter 2 addresses the mentioned research aims by reviewing the status of the
existing literature on the acceptance of automated vehicle technology and identifying crucial
factors for examining road user groups’ acceptance of CACs. Chapter 3 then describes the methods
chosen to measure the acceptance of CACs from various road users’ perspectives. The survey
results thus obtained are analysed in chapter 4. Chapter 5 discusses the study findings, and chapter
6 completes this study with a conclusion.
2. Literature
2.1 Acceptance of automated vehicle technology
5
A major challenge for measuring the acceptance of automated vehicle technology (i.e. more
advanced than SAE Level 2) by users or different road user groups is the lack of automated cars
featuring SAE Level 3 and above on the roads. Consequently, measuring the acceptance of CACs
in surveys based on objective criteria, observable behaviour or subjective experiences is not
currently possible. At present, studies are limited to approximating acceptance through
5
Although the survey and the presented results explicitly focus on automation at SAE Level 3, the following
chapter uses the term ‘automated’ for SAE Levels 3 to 5 without detailing the specific level.

EJTIR 21(4), 2021, pp.81-103 84
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
experiments, simulations or surveys examining the willingness to buy, the attitudes towards
automated cars or the intention to use them (Adell, Nilsson & Várhelyi, 2014).
6
In related research, several theoretical psychological models have been applied to predict the
individual behavioural intention to use or buy a CAC as proxies for an a priori acceptance. For
example, the Theory of Planned Behaviour (TPB) (Ajzen, 1991) and the Technology Acceptance
Model (TAM) (Davis, Bagozzi & Warshaw, 1989) or the Unified Theory of Acceptance and Use of
Technology (UTAUT) (Venkatesh et al., 2003) have been used in research investigating the
acceptance of automated vehicles (Kaye et al., 2020; Madigan et al., 2017; Nordhoff et al., 2020). In
these theoretical models, acceptance was originally recorded as the observable behaviour
concerning the use of the technical artefact. In case of the acceptance of automated cars, any
research objectives are limited to measuring behavioural intention. Different factors in these
theoretical models facilitate explaining attitudes and behavioural intentions towards automated
vehicles, such as perceived usefulness or perceived ease of use of automated vehicles (Choi & Ji,
2015; Zhang et al., 2019; for a review, see also Jing et al., 2020).
Besides these theoretically oriented studies, the body of research regarding the public acceptance
of automated vehicles is continuously growing, using respondents’ attitudes, behavioural
intentions or the willingness to pay or buy to measure this phenomenon (Becker & Axhausen, 2017;
Cunningham et al., 2019b; Daziano, Sarrias & Leard, 2017; Gkartzonikas & Gkritza, 2019; Jing et
al., 2020; Nordhoff et al., 2019). In general, empirical studies have reported mostly positive
attitudes towards the new automated vehicle technologies – or rather, positive intentions to use an
automated vehicle. For example, Schoettle and Sivak (2014a) found that 65.8% of their respondents
from the USA, UK and Australia expressed at least a slight interest in owning or leasing an SAE
Level 4 automated car. Other empirical studies, however, have indicated a certain reluctance
towards the adoption of CACs. In the BRAVE population survey, 39.0% of the respondents agreed
with the item ‘I think I will not use conditionally automated cars when available’ and only 33.2%
of the respondents expressed their future intention to use a CAC by refusing this item; 27.8% of the
respondents stayed neutral (Schrauth et al., 2020). Further empirical results have shown that
potential users would pay less for SAE Level 3 automated vehicles than for vehicles featuring SAE
Level 4 (Bansal, Kockelman & Singh, 2016; Kyriakidis, Happee & Winter, 2015). Furthermore,
acceptance studies often took a country-specific perspective and documented results e.g. for
Australia (Cunningham et al., 2019a) or Austria (Wintersberger, Azmat & Kummer, 2019). But
there is also a substantial number of studies that conduct multi-country surveys which often report
diverging results between the countries (Kaye et al., 2020; Kyriakidis, Happee & Winter, 2015;
Nordhoff et al., 2020; Schoettle & Sivak, 2014b).
Research findings also demonstrated that sociodemographic characteristics affect people’s
attitudes or behavioural intentions toward automated vehicles; for example, males tend to show a
higher acceptance of automated vehicles than females (Choi & Ji, 2015; Cunningham et al., 2019a;
Hohenberger, Spörrle & Welpe, 2016; Hulse, Xie & Galea, 2018; Kyriakidis, Happee & Winter, 2015;
Zhang et al., 2019). Although findings based on age vary or are not statistically significant in
multivariate settings, results generally tend to indicate higher acceptance among younger
respondents (Bansal, Kockelman & Singh, 2016; Nordhoff et al., 2019; Schoettle & Sivak, 2014a).
Furthermore, people living in urban areas apparently display a higher acceptance of automated
vehicle technology (Hudson, Orviska & Hunady, 2019; Piao et al., 2016). Findings reflecting
individual socioeconomic background are scarce, though some have indicated positive influence
of higher education on acceptance (Hudson, Orviska & Hunady, 2019) or of higher income on the
willingness to pay for automated technology (Bansal, Kockelmann & Singh, 2016; Kyriakidis,
Happee & Winter, 2015). In addition to individual sociodemographic characteristics, individual
attitudes toward the use of new technologies also show a positive influence on the acceptance of
automated vehicle technology (Deb et al., 2017; Zmud & Sener, 2017) or on the intention to use or
6
For a discussion on a priori acceptance and acceptability, see Adell, Várhelyi, and Nilsson (2014).

EJTIR 21(4), 2021, pp.81-103 85
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
buy it (Bansal, Kockelmann & Singh, 2016). In several studies that capture both sociodemographic
and attitude-based variables it appears that attitudinal measures outweighed characteristics such
as gender or age. Attitude-based measures often demonstrated a stronger correlation with the used
measure for the intention to use, the willingness to buy a vehicle with automated technology or
acceptance of automated vehicles in multivariate settings (Cunningham et al., 2019b; Kaye et al.,
2020; Payre, Cestac & Delhomme, 2014).
Parallel to the research on acceptance of automated vehicle technology, numerous studies have
collected insights on concerns, needs and expectations raised by the foreseeable introduction of
CACs (König & Neumayr, 2017; Schoettle & Sivak, 2014a). Gkartzonikas and Gkritza (2019) in their
meta-study provided a concise overview of extant findings. Among the frequently mentioned
benefits leading to increased acceptance of automated vehicles, the authors enumerated
improvements in road safety, increased productivity, higher fuel efficiency, lower emissions and
cost reductions for fuel, parking or insurance. Additionally, automated technology might provide
enhanced mobility for persons with reduced mobility (Piao et al., 2016). On the other hand, research
studies have uncovered serious concerns about automated technology, including various aspects
of safety issues related to system failures, legal liability or data privacy as well as the cybersecurity
of automated vehicles (Gkartzonikas & Gkritza, 2019).
2.2 Trust as a precondition for acceptance
Trust in vehicle technology, which was not explicitly taken into account in the original versions of
psychological acceptance models, was later integrated, e.g. in a revised version of the TAM
(Ghazizadeh et al., 2012). Nevertheless, researchers regard trust as a vital factor in the acceptance
of automated vehicle technology (Hoff & Bashir, 2015; Lee & See, 2004). Lee and See (2004) captured
trust as an attitude and a prerequisite for the development of behavioural intentions and actual
behaviour, making it an essential issue for measurement in the context of automated vehicle
technology. The formation of trust is described as an experience-based variable (Hoff & Bashir,
2015) which can hardly be depicted as such in current research on CACs. Thus, Hoff and Bashir
(2015, p. 420) referred to initial trust as ‘trust prior to interacting with a system’, in contrast to the
dynamic learned trust formed via interaction with a system.
Several relevant studies have proven trust being a significant factor in research on the experimental
or simulated use of automation technology (Gold et al., 2015; Hergeth et al., 2016; Payre, Cestac &
Delhomme, 2016). Other empirical studies have included this dimension in analysing the
acceptance of automated cars and verified its importance (Choi & Ji, 2015; Kaur & Rampersad,
2018; Jing et al., 2020; Zhang et al., 2019). Results indicate that higher levels of trust in the
automation technology correlate specifically with an increased intention to use automated vehicles
or, more generally, with higher acceptance. Likewise, Zmud and Sener (2017) reported a lack of
trust as the top reason for a low probability of using automated vehicles.
2.3 Acceptance of other road users
A major part of research to date has examined users' or public acceptance of automated vehicle
technology. However, this perspective overlooks the various road users who will be directly
affected and forced to interact with automated cars in road traffic. Even though many studies have
asserted that automated vehicle technology will increase road safety, for example, by reducing
human errors and following traffic rules, researchers also pointed out numerous barriers and
concerns about the pending introduction of automated vehicles that refer to users and other road
users (Gkartzonikas & Gkritza, 2019; Johnsen et al., 2018; König & Neumayr, 2017).
To date, only a few empirical studies have focused on the perspectives of other road users and
examined their attitudes towards automated vehicles (Deb et al., 2017; Penmetsa et al., 2019;
Pyrialakou et al., 2020). Deb et al. (2017) conducted a survey to identify pedestrians’ receptivity of
fully automated vehicles. Their findings indicated a positive correlation between the acceptance of
vehicle technology and three identified factors: the safe operation of an automated vehicle, the

EJTIR 21(4), 2021, pp.81-103 86
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
interaction with an automated car in a crossing situation and the compatibility of an automated
vehicle with the existing traffic system. Penmetsa et al. (2019) reported that pedestrians and cyclists
who had experienced interactions with automated vehicles on Pittsburgh’s public proving ground
expressed more positive beliefs regarding their safety than others with no interaction experience.
Pyrialakou et al. (2020), in a study involving interactions on a public testing ground in Phoenix-
Mesa-Scottsdale, Arizona, found that respondents felt safer driving near an automated vehicle than
walking or cycling close to it. In these studies, males showed more positive attitudes towards
automated vehicles than females. Moreover, the studies’ results partially confirmed the previously
mentioned findings relating to age and location from other acceptance studies.
Behavioural studies evaluating the interaction of road users with automated vehicles have
additionally pointed to possible adverse reactions of other road users toward automated vehicles
on the road. Specifically, road users might exploit the risk-averse and rule-compliant driving style
of automated vehicles by taking the right of way or driving aggressively, knowing that the
automated vehicle will stop or give way in any case (Camara et al., 2018; Liu et al., 2020; Millard-
Ball, 2018).
3. Methodology
3.1 Study design
Data from an international population survey were used to pursue the investigation of the study
aims outlined in section 1.2. The survey, which was conducted within the context of the research
project BRAVE (Schrauth et al., 2020), took place in the seven countries of the participating project
partners (the EU countries France, Germany, Slovenia, Spain and Sweden plus Australia and the
USA) and lasted from December 2019 to February 2020. The survey was conducted via computer-
assisted web interviews. Participant recruitment for the online survey was carried out using the
online access panel from KANTAR Lightspeed. The Ethics Commission of the School of Business,
Economics and Society of the University of Erlangen-Nuremberg approved this research project.
The survey was performed concurrently in all seven countries in the respective national language,
and quotas for gender, age and regions within each of the countries were applied (for more details
see Schrauth et al., 2020). In each of the participating countries, 1,000 road users aged 18 and above
were interviewed. Of these 7,000 respondents, 392 cases were deleted after data quality checks,
leaving 6,608 respondents in the data set of the BRAVE population survey for further analyses.
3.2 Questionnaire
In the BRAVE project, the questionnaire for the population survey was developed according to the
results from a preceding literature review (Johnsen et al., 2018) and explorative focus group
discussions conducted in four countries – Germany, Slovenia, Spain and Sweden (Kraetsch et al.,
2019). The questionnaire’s introduction informed the respondents about the concept of SAE Levels
as well as the characteristics of vehicles with SAE Level 3 technology and specified that the survey
questions throughout referred to CACs featuring SAE Level 3. The questionnaire covered multiple
topics and contained questions regarding the acceptance of and trust in CACs as well as concerns
and expectations towards the new automated vehicle technology. Furthermore, it included
questions exploring respondent’s affinity for technology, preferences for external human-machine
interfaces (HMIs) and the ethical and legal implications of CACs. The gathered survey data also
yielded information on preferences related to the main transportation mode, the mobility
behaviour and social demographics of the respondents. The complete questionnaire containing 62
questions and statements to assess can be viewed in the documentation of the openly accessible
data (Schrauth et al., 2021).
At the core of the questionnaire, respondents were provided with a description of a predefined
traffic situation including a fictitious interaction with a CAC in automated mode. This description
varied for the different types of the main transportation mode, including pedestrian traffic (see

EJTIR 21(4), 2021, pp.81-103 87
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
Table 1). The respondents were given the traffic situation that corresponded to their previously
indicated main transportation mode most frequently used in a working week.
7
This approach
allowed exploiting respondents’ experiences regarding their main transportation mode in the
survey. Furthermore, possible differences between road user groups could be detected. All the
presented fictitious traffic situations had in common that they contained a crossing situation in
which the respondent – as a pedestrian, cyclist, PTW rider or car driver – would have the right of
way over the CAC. Following the description of the road user-specific traffic situation, respondents
were asked to document their subjective feelings about their personal road safety in this
hypothetical situation, along with their acceptance of and trust in CACs from their point of view.
Table 1. Specified traffic situations for different road user groups
Main mode of transportation
Specified traffic situation
Pedestrian
You are walking in an urban area and want to cross the road at a pedestrian
crossing without traffic lights. At the same time, a conditionally automated
car (SAE Level 3) approaches the pedestrian crossing. The car is driving in
automated mode.
Cyclist
You are riding a bicycle in an urban area and approach a junction without
road signs or traffic lights. From the left, a conditionally automated car (SAE
Level 3) approaches. The car is driving in automated mode. You have the
right of way in this situation.
PTW rider
You are riding a powered two-wheeler in an urban area and approach a
junction without road signs or traffic lights. From the left, a conditionally
automated car (SAE Level 3) approaches. The car is driving in automated
mode. You have the right of way in this situation.
Car driver
You are driving a non-automated car in an urban area and approach a
junction without road signs or traffic lights. From the left, a conditionally
automated car (SAE Level 3) approaches. The car is driving in automated
mode. You have the right of way in this situation.
3.3 Measures
Road user acceptance
Setting up and analysing a measure for the a priori acceptance of CACs from the perspective of
other road users was one central aim of this study. As outlined earlier, a fundamental problem in
measuring the acceptance of CACs is the non-market ready stage of their technological
development and the resulting lack of approval for road traffic. The current measurement of
acceptance can only refer to road users’ a priori acceptance since the respondents most likely have
not had any experience with automated vehicles from SAE Level 3 onwards. Another issue
impedes the measurement of non-user’s acceptance: While most of the existing research has
examined the potential users’ acceptance of automated vehicle technology, comparable studies and
theoretical concepts for developing and assessing the measurement tool for road user acceptance
proposed in this study are hardly available.
The approach chosen in the BRAVE population survey was based on the findings of
psychologically oriented user-centred research (Johnsen et al., 2018). Existing research on user
acceptance employing simulations or experiments has already identified dimensions that provide
explanatory power for predicting user acceptance, such as perceived ease of use, perceived
usefulness or the compatibility of (conditionally) automated vehicles (Choi & Ji, 2015; Ghazizadeh
et al., 2012; Jing et al., 2020; Zhang et al., 2019). These dimensions were used for an attitude-based
7
The use of public transport was omitted because users of public transport would not be in direct interaction
with the CAC. Instead, respondents were asked to consider the way to or from public transport in indicating
their main transportation mode.

EJTIR 21(4), 2021, pp.81-103 88
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
measurement of road users’ acceptance of CACs (see Table 2). Accordingly, the three items listed
in Table 2 were specially (re-)formulated or modified according to the prevailing intention of the
population survey. In the questionnaire, these items followed the description of the traffic scenario
outlined in section 3.2 and their formulation was individually adapted to reflect the respondents’
indicated main mode of transportation so that the specific perspective of a pedestrian, cyclist, PTW
rider or driver of a non-automated car was collected. Respondents could state their opinion on a 5-
point Likert scale ranging from 1 ‘I strongly disagree’ to 5 ‘I strongly agree’.
Table 2. Items for the measurement of road user acceptance of CACs and descriptive
statistics
Item
Dimension
n
x
SD
Cron-
bach’s 
As a [pedestrian/cyclist/rider of
a PTW/driver], I think
conditionally automated cars will
be easy to communicate with.
Perceived ease of use
Elaboration based on:
Ghazizadeh et al. (2012)
5,827
3.20
1.05
0.78
As a [pedestrian/cyclist/rider of
a PTW/driver], I think that
conditionally automated cars will
cause problems for me and other
road users.8
Compatibility
Elaboration based on:
Ghazizadeh, Lee, and Boyle
(2012)
5,827
3.15
1.05
As a [pedestrian/cyclist/rider of
a PTW/driver], I think that
conditionally automated cars will
make roads safer.
Perceived usefulness
Elaboration based on:
Ghazizadeh et al. (2012)
5,827
3.30
1.03
* Response categories for items on road user acceptance: 1 = ‘I strongly disagree’, 2 = ‘I disagree’, 3 = ‘I neither
agree nor disagree’, 4 = ‘I agree’, 5 = ‘I strongly agree’.
* SD = standard deviation
To assess the consistency of the scale, the reliability measure of Cronbach's alpha is used.
Cronbach’s alpha for the three items was 0.78, which is above the generally agreed lower limit of
0.70 and thus allows further use in one index (Hair et al., 2014). Hence, these three items were used
to calculate an additive index representing the a priori road user acceptance of CACs. The index
was calculated by adding the values of each of the three items and dividing the sum by the number
of items resulting in an index that ranges from 1 to 5 corresponding to the single items. The built
index reflects the measurement of an a priori acceptance of CACs from the perspective of road
users and forms the central target variable of bi- and multivariate analyses in this study.
9
General trust in CACs
The measurement of general trust in CACs was carried out with an already existing scale consisting
of three items and having proven reliability (Choi & Ji, 2015; Zhang et al., 2019). As with the
measure of acceptance, these items cannot depict trust formed by real experiences but can represent
the initial trust that road users may place in CACs (Hoff & Bashir, 2015). The formulation of the
items has been adapted to the context of CACs. For further use in statistical analyses, the three
items were combined into an additive index in the same way as for the road user acceptance before.
The reliability analysis for the three items yielded a Cronbach’s alpha of 0.89 (see Table 3).
8
Because of the negative formulation of this item in contrast to the other two items on road user acceptance, the
values of this variable were recoded for the calculation of the additive index of the road user acceptance. With
the reversed coding, the values of all items now point in the same direction where higher values correspond
with more positive attitudes towards CACs.
9
If not otherwise emphasised in the following text, the road user acceptance in this study always refers to an a
priori measurement.

EJTIR 21(4), 2021, pp.81-103 89
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
Personal innovativeness
Several studies have shown the importance of enthusiasm to use or try out technology, e.g. Deb et
al. (2017) identified personal innovativeness as a predictor for the acceptance of automated vehicle
technology. Innovativeness describes a personal interest in testing and using new technical
devices. In this study, an already existing measurement instrument was used to assess personal
innovativeness (Agarwal & Prasad, 1998) that has also been used in Deb et al. (2017). For the
questionnaire, three items were selected from the original instrument and slightly modified to fit
the research subject. To conduct further statistical analyses, an index was formed out of the three
items. Again, the additive index was formed in the same way as the index of the road user
acceptance before. Cronbach’s alpha for these three items was 0.75 (see Table 3).
Table 3. Items and descriptive statistics for measurement of general trust in CACs and
personal innovativeness
Item
n
x
SD
Cronbach’s 
General trust in CACs
Conditionally automated cars will be dependable.
5,827
3.34
1.02
0.89
Conditionally automated cars will act reliably.
5,827
3.31
1.02
Overall, I will trust conditionally automated cars.
5,827
3.20
1.12
Personal innovativeness
Among my peers, I am usually the first to try out
new technologies.
5,827
2.83
1.18
0.75
In general, I am hesitant to try out new
technologies.10
5,827
2.74
1.16
I like to experiment with new technologies.
5,827
3.54
1.09
* Response categories for items on general trust in CACs and personal innovativeness: 1 = ‘I strongly disagree’,
2 = ‘I disagree’, 3 = ‘I neither agree nor disagree’, 4 = ‘I agree’, 5 = ‘I strongly agree’.
Sociodemographics and mobility behaviour
Next to the attitude-based measures, several information on sociodemographics and the mobility
behaviour of the road users surveyed are available in the dataset. The survey data entails
information on gender and age as well as measures for the educational level, the socioeconomic
status and the place of living. The measurement of these latter three variables follows standardized
questions from the International Social Survey Programme (ISSP), which ensures international
comparability (GESIS Leibniz Institute for the Social Science, 2019). The educational level was
measured by the highest educational degree following a slightly modified standardised procedure
specified by the ISSP for each country (ISSP Research Group, 2019). The socioeconomic status was
depicted with an existing self-placement scale for social positioning within the society. This
instrument measured the self-assessment of respondents using a 10-point scale from 1 ‘Lowest,
Bottom’ to 10 ‘Highest, Top’. The respondents’ place of living was operationalized using five
categories of which the two categories ‘village’ and ‘farm’ were combined due to the low number
of respondents on farms.
Measures reflecting the mobility behaviour of the road users surveyed were the average frequency
of trips on a day from Monday to Friday, the most frequently used mode of transportation for
everyday private mobility and the previous experience with advanced driver assistance systems
(ADAS) which was measured on a 4-point Likert scale ranging from 1 ‘Never’ to 4 ‘Often’.
10
Like in footnote 8, the negative formulation of this item was considered in data analysis by recoding the values
of the variable.

EJTIR 21(4), 2021, pp.81-103 90
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
3.4 Data analysis
Data analysis was performed in two steps. In the first step, the results of the measurement tool
determining road user acceptance were examined with the use of univariate and bivariate analysis
methods. The bivariate analysis used a set of independent variables, including gender, age,
respondents’ country of residence and main transportation mode demonstrating underlying
correlations of the variables with the road user acceptance. In the second step, a multivariate
analysis was applied, considering the calculated additive index of road user acceptance as the
dependent variable. By using a multivariate analysis method, the individual effects of the different
determinants can be depicted in isolation while controlling for the chosen set of independent
variables. According to the metric level of the dependent variable, a two-step linear regression was
carried out. From the 6,608 respondents remaining in the original sample of the BRAVE population
survey, some were removed due to missing values mostly caused by item non-response, leaving
5,827 records for the univariate and bivariate as well as for the multivariate analyses.
11
The
predictors of age and respondents’ socioeconomic status as well as personal innovativeness and
general trust in CACs are included as metric variables in the regression analysis. The categorical
independent variables of main mode of transportation, place of living, highest educational level,
number of trips per day, and experience with ADAS are added to the model in dichotomized form.
So is gender, which is also included dichotomously in the calculation. The analytical strategy took
into account the location of the respondents in the individual countries to adequately address the
multi-level research design of the survey, which had been conducted in seven countries. Thus, the
multivariate analysis used one common model for all countries with country-specific clustered
standard errors (Bryan & Jenkins, 2016). Data analysis was performed using Stata IC 13.
4. Results
4.1 Demographics and technology-related attributes
In the data set prepared for the data analyses (n = 5,827), 50.0 % female road users remained. On
average, the survey participants had an age of  = 44.69 years (standard deviation SD = 16.44).
Regarding the respondents’ educational level, the category of educational degrees on tertiary level
formed the largest group with 45.2%, participants with upper or post-secondary degrees comprise
39.5%. The lowest proportion was made up by road users with lower or no educational degrees
(15.3%). Regarding the self-reported estimation of socioeconomic status, the respondents ranked
themselves, on average, somewhat above the middle ( = 6.23; SD = 1.66).
26.1% of road users in the dataset lived in cities and 23.0% in suburbs. The largest proportion of
31.1% resided in towns and 19.2% of the road users stated that they live in villages or on farms.
Regarding the mode of transport used most often for everyday private mobility, car drivers formed
the largest (66.9%) and pedestrians the second largest proportion (24.7%); cyclists (6.0%) and
motorcyclists (2.4%) were in the minority. The median number of trips per day was  = 3.
Of the road users surveyed, 15.5% had already often used ADAS up to the time of the survey.
Furthermore, in the survey 25.2% stated they have sometimes used ADAS and 22.2% stated that
they have rarely done so. The share of respondents not having used ADAS was the largest with
37.1%. Regarding the respondents’ enthusiasm of dealing with new technologies, the index for
personal innovativeness yielded an average of  = 3.21 (SD = 0.93) and the average general trust in
CACs was  = 3.28 (SD = 0.95).
4.2 Univariate analysis of items regarding road user acceptance
The additive index calculated from the three items introduced in Table 2 yielded an average value
of  = 3.12 (SD = 0.87) on a scale ranging from 1 ‘I strongly disagree’ to 5 ‘I strongly agree’. The
11
The results reported below may differ from results presented in Schrauth et al. (2020), as this paper’s findings
refer to the cases included in the multivariate analysis only.

EJTIR 21(4), 2021, pp.81-103 91
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
response frequencies to the three items underlying the formation of road user acceptance are
presented in Table 4. 44.7% of the road users in the sample expected easy communication with
CACs, remarkably less survey participants (strongly) disagreed with this statement (27.1%). The
statement that CACs will make roads safer was (strongly) agreed to by 47.1%. This assessment is
contradicted by 23.2% of the road users surveyed. In contrast to the rather positive attitudes before,
39.7% of road users expected problems for themselves and other road users; 30.1% of the
respondents, on the opposite, did not expect any problems. The univariate findings suggest a basic
acceptance of CACs in which positive agreement exceeds disapproval regarding the
communication with the CAC and the expected improvements in road safety. However, the
findings simultaneously signal doubts about how the new technology will fit into the current road
traffic system.
Table 4. Items and descriptive statistics for measurement of the road user acceptance of
CACs
Item
n
I strongly
agree
I agree
I neither
agree nor
disagree
I dis-
agree
I strongly
disagree
%
As a [pedestrian/cyclist/rider of a
PTW/driver], I think conditionally
automated cars will be easy to communicate
with.
5,827
8.4
36.3
28.3
21.3
5.8
As a [pedestrian/cyclist/rider of a
PTW/driver], I think that conditionally
automated cars will make roads safer.
5,827
10.2
36.9
29.8
18.6
4.6
As a [pedestrian/cyclist/rider of a
PTW/driver], I think that conditionally
automated cars will cause problems for me
and other road users.
5,827
9.5
30.2
30.3
25.6
4.5
4.3 Bivariate analysis of the road user acceptance
For subsequent bivariate analyses, the additive index of the road user acceptance of CACs was
used as the dependent variable. A correlation analysis of road users’ acceptance and gender using
Pearson’s product-moment correlation coefficient resulted in a moderate negative correlation
(r = -0.13, p < 0.001). This result suggests that the female road users demonstrated less road user
acceptance of CACs than males. A negative correlation also resulted from the bivariate correlation
analysis between road user groups’ acceptance and age (r = -0.20, p < 0.001), signalling that
acceptance of CACs decreased with advancing age. Categorising age into subgroups as presented
in Table 5, furthermore, a non-linear relationship became apparent saying that in particular the
oldest age group demonstrates a lower acceptance of CACs.
Road user acceptance differed greatly depending on the respondents’ countries of residence
(Kruskal-Wallis H test with χ2(6) = 220.041, p < 0.001; Cramer’s V = 0.10, p < 0.001). Respondents from
Spain and Slovenia showed, on average, the highest acceptance of CACs (see Table 5). Meanwhile,
respondents from Sweden, Australia and France ranked midfield, and road users from the USA
and Germany comparably reported the lowest acceptance of CACs on the street.
The subgroup analysis by the means of transport most frequently used in a working week also
demonstrated statistically significant differences (Kruskal-Wallis H test with χ2(3) = 21.214, p < 0.001;
Cramer’s V = 0.06, p < 0.01). Among the road user groups, PTW riders showed the highest
acceptance of CACs on the road, followed by car drivers and cyclists (see Table 5). In all countries,
pedestrians indicated the lowest acceptance of CACs on the road. That is, motorized road user
groups showed higher acceptance of CACs than cyclists and pedestrians.

EJTIR 21(4), 2021, pp.81-103 92
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
Table 5. Index of road user acceptance by gender, age, respondents’ country of residence and
main mode of transportation
Variable
n
x
SD
Index of road user acceptance
5,827
3.12
0.87
By gender
Female
2,913
3.01
0.84
Male
2,914
3.23
0.88
By age (in years)
Up to 24
521
3.21
0.78
25 to 34
1,473
3.27
0.80
35 to 44
1,330
3.23
0.84
45 to 54
812
3.19
0.87
55 and more
1,691
2.84
0.90
By country
France
832
3.03
0.79
Germany
866
2.97
0.89
Slovenia
818
3.36
0.78
Spain
853
3.38
0.80
Sweden
817
3.10
0.85
Australia
832
3.06
0.86
USA
809
2.92
0.98
By main mode of transportation
Pedestrian
1,441
3.03
0.86
Cyclist
350
3.11
0.83
PTW rider
138
3.22
0.66
Car driver
3,898
3.15
0.88
Bivariate analyses with the attitude-based predispositions of general trust in CACs (= 3.28,
SD = 0.95) and personal innovativeness ( = 3.21, SD = 0.93) revealed a strong relationship with
the index of road user acceptance. The calculated correlation between road user group acceptance
and general trust was r = 0.71 (p < 0.001). As proposed by the theoretical models, this strong
positive correlation highlights the essential role of trust for the acceptance of CACs. The correlation
coefficient for the relationship between road user acceptance and personal innovativeness was
r = 0.46 (p < 0.001). Thus, personal interest to experiment with new technologies showed a
moderately strong positive correlation with the acceptance of CACs.
4.4 Multivariate analysis of the road user acceptance
This section reports the results of a multivariate two-step linear regression analysis carried out
with the index of road user acceptance of CACs as the dependent variable. In the first step of the
multivariate analysis, the relationships between the dependent variable and variables covering
sociodemographic characteristics such as gender and age
12
, the main transportation mode, the
frequency of daily trips and experience with ADAS were estimated (see Table 6). Likewise, the
impact of the location of residence, the socioeconomic status and the highest educational level of
the respondents on the specific acceptance of CACs was examined in the analysis since the extant
research results have emphasised their relevance. The second step additionally included the
variables of attitude-based predispositions ‘personal innovativeness’ and ‘general trust in CACs’
in Model 2.
12
In the regression, age was included as a metric variable. The addition of age squared addressed the curvilinear
relationship indicated in the preceding bivariate analyses (see Table 5).

EJTIR 21(4), 2021, pp.81-103 93
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
Table 6. Linear regression models with road user acceptance of CACs as the dependent
variable; use of robust standard errors, clustered by country (i.e. standard errors
(SE), confidence intervals (CI))
Variable
Model 1
Model 2
Unstandardised
Coefficient
Robust SE
CI
Unstandardised
Coefficient
Robust SE
CI
Female (ref.: male)
-0.21***
0.03
-0.29, -0.13
-0.05*
0.02
-0.09, -0.00
Age
0.01
0.01
-0.01, 0.03
0.01
0.00
-0.00, 0.02
Age2
-0.00
0.00
-0.00, 0.00
-0.00
0.00
-0.00, 0.00
Main mode of transportation
(ref.: car driver)
Pedestrian
-0.09*
0.03
-0.17, -0.01
-0.10**
0.02
-0.15, -0.05
Cyclist
-0.14*
0.05
-0.25, -0.02
-0.14*
0.04
-0.24, -0.05
PTW rider
-0.07
0.08
-0.26, 0.13
-0.21*
0.07
-0.37, -0.05
Location of residence
(ref.: farm/village)
Town
0.07
0.05
-0.05, 0.20
0.02
0.04
-0.07, 0.12
Suburb
0.05
0.04
-0.06, 0.15
-0.03
0.03
-0.11, 0.04
City
0.16
0.07
-0.01, 0.34
-0.01
0.03
-0.08, 0.07
Highest level of education
(ref.: low educational level)
Middle educational level
-0.02
0.04
-0.13, 0.08
-0.01
0.02
-0.07, 0.05
High educational level
-0.02
0.03
-0.09, 0.06
-0.03
0.02
-0.08, 0.02
Social positioning
0.04**
0.01
0.02, 0.07
-0.01
0.01
-0.02, 0.01
Average number of trips per day
(ref.: none or one trip per day)
Two trips per day
0.09
0.07
-0.07, 0.26
0.03
0.05
-0.09, 0.14
Three to four trips per day
0.14
0.06
-0.01, 0.28
0.03
0.04
-0.08, 0.13
Five to eight trips per day
0.13*
0.05
0.010, 0.26
0.05
0.04
-0.05, 0.15
Nine and more trips per day
0.13
0.07
-0.05, 0.31
0.04
0.06
-0.10, 0.18
Experience with ADAS (ref.: never)
Rarely
0.22***
0.02
0.17, 0.28
0.01
0.03
-0.06, 0.08
Sometimes
0.33***
0.32
0.26, 0.41
0.03
0.02
0.00, 0.07
Often
0.41**
0.06
0.26, 0.56
0.03
0.03
-0.04, 0.09
Personal innovativeness
0.17***
0.01
0.15, 0.20
General trust in CACs
0.57***
0.02
0.52, 0.63

EJTIR 21(4), 2021, pp.81-103 94
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
Table 6. (continued)
Variable
Model 1
Model 2
Unstandardised
Coefficient
Robust SE
CI
Unstandardised
Coefficient
Robust SE
CI
Constant
2.61***
0.16
2.21, 3.00
0.53*
0.17
0.12, 0.94
Number of observations
5,827
5,827
Adjusted R-square
0.12
0.54
* Significance level: * p < 0.05, ** p < 0.01, *** p < 0.001.
Table 6 presents the results of the linear regression analysis, including 5,827 respondents. Model 1
achieved an explanatory power of Adjusted R2 = 0.12 indicating an only weak explanatory power.
Statistically significant predictors in this model were the road users’ gender and belonging to the
vulnerable road user groups of pedestrians and cyclists – each in contrast to the drivers of
conventional cars. Despite the significance in bivariate analysis, age and age squared did not yield
a significant prediction in the regression model. Further variables of statistical relevance were the
self-reported social positioning, a certain frequency of trips per day (five to eight) and previous
experience with ADAS. These three variables all exerted a positive influence on road user
acceptance in the first regression model. In other words, a higher self-placement on the ten-point-
scale of the social positioning towards upper social strata, a higher number of trips per day, and
more experience with ADAS significantly correlated with higher acceptance of CACs. In contrast,
being female lowered road user acceptance as well as being mainly a pedestrian or a cyclist in
reference to driving a conventional car. Other sociodemographic factors such as age or educational
level did not yield a statistically significant relationship in the regression model.
The introduction of the attitude-based predispositions ‘personal innovativeness’ and ‘general trust
in CACs’ in Model 2 remarkably increased the predictive power of the statistical regression model
(Adjusted R2 = 0.54). Simultaneously, some of the observed results in Model 1 disappeared, or the
calculated strength of the correlation decreased in the second model. Of the variables identified in
Model 1, gender and the main mode of transportation remained statistically significant predictors
for the acceptance of CACs. Though, the strength of the correlation between gender and the
dependent variable was considerably reduced in Model 2, females continued to exhibit
significantly less acceptance of CACs than males. In contrast, the relationship between the main
mode of transportation and the acceptance of CACs became more prominent and stronger in
Model 2. Compared to drivers of manually driven cars, pedestrians, cyclists and PTW riders
demonstrated significantly lower road user acceptance levels. This indicates that the modes of
transportation attributed to VRUs stood out markedly in their lower acceptance of CACs. Whereas
drivers of conventional cars, on the opposite, showed the greatest acceptance of CACs in the
fictitious traffic scenarios in the questionnaire presented.
In Model 2, the indices of personal innovativeness and general trust in CACs proved statistically
significant and appeared to be positive and rather strong determinants of road user acceptance in
CACs. Comparatively, general trust in CACs showed an even stronger positive correlation than
personal innovativeness. According to these results, the coefficients of the two attitudinal variables
in the second model and the additional explanatory power show that personal innovativeness and,
in particular, general trust in CACs were essentially related to road user acceptance of CACs as
measured in the predefined traffic scenarios.

EJTIR 21(4), 2021, pp.81-103 95
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
5. Discussion
5.1 Synthesis of findings
This study assessed the a priori acceptance of CACs from the perspective of different road user
groups. Upon the market-wide introduction of CACs, road users such as pedestrians, cyclists and
PTW riders as well as drivers of non-automated cars will be forced to interact with CACs, raising
questions about their safety, especially among VRUs. Accordingly, the first aim of this study was
to measure the acceptance of CACs from the perspective of the different road user groups.
Furthermore, as a secondary goal, the presented study explored whether the road user groups
differed in their acceptance of CACs on SAE Level 3. These research aims were investigated with
a specifically developed measurement tool, building on previous research on user-centred
acceptance of automated vehicle technology. Given prior findings, the items of this tool refer to the
dimensions of the perceived ease of use, the compatibility with prevailing road traffic and the
perceived usefulness and are adapted to the perspective of the road users. The questionnaire of the
BRAVE population survey initially presented a description of a specific traffic situation involving
an interaction with an approaching CAC. This description differed according to the individually
used main mode of transportation, followed by the three items measuring the road user’s
acceptance of CACs.
The univariate results of the items for road user acceptance indicated a basic acceptance of CACs
in which communication with the CACs and their benefits for road safety were rated quite
positively. In this regard, the findings basically correspond with previous studies on public or user
acceptance of automated vehicle technology (Cunningham & Regan, 2020; Jing et al., 2020).
Compared to the positive assessment towards communication with CACs and their road safety,
however, findings revealed doubts from the road users about the compatibility with the existing
conditions in road traffic. Therefore, concerns raised regarding the reliability, cybersecurity or
liability of CACs as well as detecting the intentions of other road users in the BRAVE population
survey (Schrauth et al., 2020) should be thoroughly addressed in the further development and
implementation of automated vehicle technology to further improve acceptance among all road
user groups. An introduction of SAE Level 3 vehicles with insufficient acceptance among road
users might lead to rejection or undesired behaviour of other road users towards automated
vehicles, e.g. bullying or exploiting the risk-averse driving style of automated cars (Camara et al.,
2018; Liu et al., 2020).
Results on sociodemographic variables from the bi- and multivariate analyses performed with the
additive index of the road user acceptance of CACs recall similar patterns in familiar research on
user or public acceptance (Hulse, Xie & Galea, 2018; Kyriakidis, Happee & Winter 2015; Nordhoff
et al., 2020). Regarding gender, female road users showed a lower specific acceptance of CACs than
males. This relationship has been statistically confirmed in bi- and multivariate analyses. Though
in Modell 2 of the regression analysis, the gender effect weakened considerably after controlling
for general trust and personal innovativeness. A weakening or non-existent gender effect after
controlling for attitude-based variables in multivariate analyses was already evident in user-
centered acceptance studies of Kaye et al. (2020) or also Payre, Cestac & Delhomme (2014).
Similarly, the predictors of age and age squared did not become statistically significant in the
regression models as they did in bivariate analyses before. Comparable findings on weak or
diminishing effects of age on the acceptance of automated vehicle technology in multivariate
settings were also reported in Nordhoff et al. (2019), Kaye et al. (2020) and again in Payre, Cestac
& Delhomme (2014). These results reflect previous findings and underline the assumption
proposed by Hohenberger, Spörrle & Welpe (2016) that attitudes may mediate the correlation
between gender or age and the acceptance of automated vehicle technology.
Besides gender and age, other sociodemographic or mobility-related predictors, such as the
frequency of trips per day, the experience with ADAS, the respondent’s place of living, the highest
level of education or subjective social positioning, were not relevant or rather lost their statistical

EJTIR 21(4), 2021, pp.81-103 96
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
significance in step two of the regression analysis. Instead, general trust and personal
innovativeness were the central determinants for explaining road user acceptance of CACs. Both
attitude-based determinants had the strongest correlation with the road user acceptance of CACs,
while gender, age and other sociodemographic as well as mobility-related variables only showed
a small, often non-significant relation with the explanandum in the multivariate setting.
These findings point out the central role of trust and innovativeness for the road user acceptance
of CACs, which largely outweighs sociodemographic or mobility-related characteristics. As in
other research, a reasonable conclusion is that trust in and enthusiasm for technology determine
the acceptance of CACs among road users (Deb et al., 2017; Zhang et al., 2019). From the present
results, it appears that trust could be one of the crucial factors to increase the acceptance of CACs
among other road users. Further studies should thus also place trust in CACs at the centre of
research interest and investigate its determinants more closely. First experiences in interaction with
automated vehicles demonstrate positive effects on building trust, as has already became apparent
in the research of e.g. Penmetsa et al. (2019) and underpinning the experienced-based character of
trust (Hoff & Bashir, 2015). Though, it is challenging building trust in a vacuum where no CACs
are yet driving in real traffic. What an appropriate trust-building measures in the current stage
could look alike requires further efforts from research, industry and politics.
Next to the development of a measurement tool for road user acceptance, the second central aim
of this study was to investigate the differences in the acceptance of CACs among distinct groups
of road users. In this respect, the findings demonstrate that the road user groups significantly
varied in their specific acceptance of CACs. In the multivariate analysis, all groups of VRUs –
pedestrians, cyclists and PTW riders – revealed a lower acceptance of CACs in contrast to the non-
automated car drivers. Interestingly, results for PTW riders in the second model reveal a stronger
negative correlation than in the first model. This effect can be explained by examining the bivariate
correlations between the main mode of transportation and general trust as well as personal
innovativeness of the different road user groups: PTW riders have above-average general trust in
CACs and above-average personal innovativeness compared to the other road user groups.
Controlling for these relations in the second model, the net effect remains for the PTW riders, who
then demonstrated a strongly reduced acceptance compared to car drivers.
According to these results, the road user groups showed diverging levels of acceptance of CACs,
with VRUs expressing greater concerns about their subjective safety than car drivers. Two reasons
seem plausible for this assessment: first, it is conceivable that VRUs assume they are less likely to
be detected by CACs than cars and anticipate coordination problems with CACs. In addition, VRUs
are less protected in the event of a crash with a CAC and perceive themselves to have a higher risk
of injury when colliding with a CAC. Additionally, it fits the picture that cyclists and PTW riders
who will share the road with CACs are even less accepting CACs than pedestrians. The interaction
of CACs with pedestrians, cyclists and PTW riders must receive strong attention during
technological development to ensure that the safety of the VRUs is not jeopardised by unsuitable
detection systems that cannot precisely predict the VRUs’ intentions.
Moreover, it is remarkable that the addition of trust and innovativeness in the second model
reduced the effect sizes of most determinants regarding variables covering sociodemographics and
mobility behaviour – but not those of the road user groups. The effect sizes and differences between
the road user groups remained persistent – or even became more explicit – and signal that besides
general trust-building measures, especially the individual road user groups must be targeted with
policies fostering acceptance of CACs.
Above that, findings from the bivariate analysis pointed out substantial differences in road user
acceptance of CACs among countries. Against the background of previous cross-national studies
on public or user acceptance, the results fit the picture of varying degrees of acceptance between
different countries (Cunningham & Regan, 2020) or different determinants within country-specific
analyses (Kaye et al., 2020). However, it is difficult to clearly attribute the differences between these

EJTIR 21(4), 2021, pp.81-103 97
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
seven western and industrialized countries neither to varying information regarding CACs in the
countries, diverging national technological progress, cultural aspects nor to the diversity in the
transport system. Similarly, Kaye et al. (2020) in their comparative study also found country-
specific results but could not clearly attribute them to specific reasons. Exemplary, Nordhoff et al.
(2020) in their international study likewise acknowledged country differences and declared this to
be still a research target. In this regard, more cross-country comparative studies would have to be
carried out to obtain a possible consistent picture of country differences. At the same time, further
multi-country studies would have to aim at exploring the effects of e.g. cultural aspects or national
conditions. A more detailed cross-country analysis of the acceptance of CACs seems essential as
the introduction of CACs will not be limited to one country only. International regulations for
manufacturing or also cross-border traffic require cross-national consensus.
5.2 Practical implications
A core result of the study involved the clear differences in the acceptance of CACs between the
distinct groups of road users. This road user group-specific acceptance was lower among VRUs
than among drivers of conventional cars. The reason for the VRUs’ more reserved acceptance might
stem from higher scepticism towards self-driving cars, possibly resulting from higher uncertainty
about interacting with these vehicles. Next to the implicit communication like the approach speed
of CACs, the installation of external HMIs on automated vehicles might be one way to reduce such
uncertainty. Examples of preferred solutions that were also surveyed in the BRAVE population
study include an indication when the CAC is in automated mode or flashing light signals at the car
when the CAC approaches a pedestrian crossing and gives way (Schrauth et al., 2020). The use of
external HMIs on automated vehicles is already being researched and tested (Rouchitsas & Alm,
2019; Schieben et al., 2019). Results, however, show that attention must be paid to a moderate use
of external HMIs to ensure comprehensible communication with road users via external HMIs
(Kaleefathullah et al., 2020). Uniform and barrier-free concepts must also be considered, as well as
the opportunity for road users to learn how to interpret the external HMI system.
Above that, communication via external HMIs as currently designed is unidirectional and
originates solely from the CAC. In certain traffic situations, the pedestrian, cyclist or PTW riders
may wish to communicate with the CAC and may wave the CAC past, for example. Such
bidirectional communication does not seem to be foreseen in the current use of external HMI and
is certainly an issue that needs to be addressed in the future.
The installation of external HMIs on CACs might also generate a trust-building effect. As the
results presented here and in other research studies show, trust will be a substantial component in
encouraging the acceptance of CACs – from the specific perspective of the different road user
groups as well as from the viewpoint of potential users. Regarding the presented results, especially
pedestrians, cyclists and PTW riders should be sensitised to interact with CACs. Likewise,
women’s scepticism about automated vehicle technology might also be addressed in
communication campaigns that are expected to increase acceptance of automated vehicles
(Pettigrew et al., 2019).
At the same time, technological developments and regulatory standards for the approval of
automated vehicles must give the studied groups and all other population groups every reason to
place trust in the technology and road safety of CACs. In the current absence of CACs in real traffic,
it seems important for policymakers and the industry to create a trust-building framework through
appropriate communication and by transparently setting the regulations for e.g. ethical and legal
issues that are in line with societal values.
5.3 Strengths and limitations
The investigation of road user acceptance of CACs was conducted within a large-scale population
survey including seven countries. Thus, the study took advantage of a commercial online access
panel that has some considerable limitations (Couper, 2017). Besides the unfavourable response

EJTIR 21(4), 2021, pp.81-103 98
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
behaviour addressed by checks for the length of interviews and the identification of indifferent
response behaviour (Schrauth et al., 2020), the representativeness of the data is another issue.
Nevertheless, comparisons of selected sociodemographic data between the survey data and official
population data from the United Nations (2019) demonstrate that the recruited survey participants
approximately reflect the population in the seven countries in terms of gender and age (Schrauth
et al., 2020). However, survey data from online access panels principally lack generalisability to the
whole population of a country and must be considered representative only for Internet users in
that country.
In the present study, an in-depth analysis of country differences was neglected in favour of
measuring factors influencing the road user acceptance of CACs; these differences were instead
considered a cluster variable in the regression. Hence, while this paper does not address the
differences between the individual countries that appeared in the bivariate analyses, they are
displayed in more detail in Schrauth et al. (2020). As already outlined, explanations for these
diverse results regarding countries have not yet been explored in detail and should be addressed
in further studies. In this regard, approaches dealing, for example, with legal regulations for a
worldwide introduction of automated vehicle technology must consider national differences,
which relate not only to road users’ acceptance but also to initial trust in conditional automated
technology, as well as technical and ethical considerations (Schrauth et al., 2020).
Moreover, the measurement tool for road user acceptance of CACs has been applied for the first
time and in the context of one specified traffic situation. In doing so, the study primarily referred
to extant findings from user-centred research on automated vehicles and draws on three central
dimensions for the formulation of its three items. Comparable research results of acceptance of
CACs from a non-user perspective are rare, making it difficult to assess the validity – however,
they appear, against the backdrop of existing findings, theoretically and empirically plausible. Of
course, this situation is also – as in all other studies – a matter of missing possibilities to measure
acceptance of or trust in the new automated vehicle technology against real conditions. Therefore,
using the measurement tool in other contexts as well would be desirable, e.g., in (quasi-)
experimental research designs to verify the differences between the road user groups found in this
study.
Another methodological remark concerns the operationalisation of the general trust in CACs and
its textual proximity to the measurement tool of the road user acceptance of CACs. In the data
analysis, it must be critically questioned, whether the strong correlation between trust and road
user acceptance may also refer to the operationalisation of both variables. Future studies may, thus,
try a different indicator measuring initial trust in CACs to better separate the two theoretical
constructs of trust and acceptance.
6. Conclusion
The present study focused on the a priori acceptance of CACs from the perspective of different
road user groups who will have to interact with automated vehicles after their introduction –
without actively having ‘decided’ to do so. Following the principles of a Pareto-optimal
improvement on the introduction of CACs, the different road users may not be put in a worse
position. For example, they must not experience a higher risk to their subjectively perceived or
objective road safety, nor must they inevitably encounter restrictions to their freedom of movement
in road traffic – especially as high expectations on automated vehicle technology ’are at present
largely speculative and yet to be proven‘ (Cunningham & Regan, 2020, p. 96).
For this reason, the legitimate interests of the different road user groups, particularly the VRUs,
must be sufficiently considered in the development of automated vehicle technology prior to a
market-wide introduction. Findings from this study reveal a basic acceptance on the part of the
road users towards interactions with CACs on the road, though concerns about such interactions

EJTIR 21(4), 2021, pp.81-103 99
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
were apparent as well. Moreover, the presented findings provide initial evidence that, at least in
the traffic situations defined in the questionnaire, the acceptance of CACs differs between the road
user groups and thereby has identified a road user group-specific acceptance of CACs.
Furthermore, it is remarkable that from their point of view, the VRUs currently feel unequally
affected by the introduction of CACs and reveal a lower acceptance of CACs than drivers of non-
automated cars.
These results provide a reason for further investigation of the acceptance of automated vehicle
technology from the perspective of different road user groups. A greater consideration could
ensure better identification of issues raised by road user group-specific requirements for
interaction and communication with CACs. Nevertheless, it must be acknowledged that in this
context, not only the requirements of the different road user groups already discussed must be
considered, but further thought should be given to the needs of children, elderly or people with
disabilities.
Acknowledgements
This paper builds upon data gathered in the research project ‘BRidging gaps for the adoption of
Automated VEhicles’ (BRAVE), which received funding from the European Union’s Horizon 2020
research and innovation program under grant agreement No 723021.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in the paper.
References
Adell, E., Nilsson, L., & Várhelyi, A. (2014). How Is Acceptance Measured? Overview of Measurement
Issues, Methods and Tools. In M. A. Regan, T. Horberry, & A. Stevens (Eds.), Driver acceptance of new
technology: theory, measurement and optimisation (pp. 73–88). Farnham Surrey, England, Burlington, VT:
Ashgate Publishing Company.
Adell, E., Várhelyi, A., & Nilsson, L. (2014). The definition of acceptance and acceptability. In M. A.
Regan, T. Horberry, & A. Stevens (Eds.), Driver acceptance of new technology: theory, measurement and
optimisation (pp. 11–21). Farnham Surrey, England, Burlington, VT: Ashgate Publishing Company.
Agarwal, R., & Prasad, J. (1998). A Conceptual and Operational Definition of Personal Innovativeness
in the Domain of Information Technology. Information Systems Research, 9(2), 204–215.
https://doi.org/10.1287/isre.9.2.204
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes,
50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T
Banks, V. A., Eriksson, A., O’Donoghue, J., & Stanton, N. A. (2018). Is partially automated driving a bad
idea? Observations from an on-road study. Applied Ergonomics, 68, 138–145.
https://doi.org/10.1016/j.apergo.2017.11.010
Bansal, P., Kockelman, K. M., & Singh, A. (2016). Assessing public opinions of and interest in new
vehicle technologies: An Austin perspective. Transportation Research Part C: Emerging Technologies, 67, 1–
14. https://doi.org/10.1016/j.trc.2016.01.019
Becker, F., & Axhausen, K. W. (2017). Literature review on surveys investigating the acceptance of
automated vehicles. Transportation, 44(6), 1293–1306. https://doi.org/10.1007/s11116-017-9808-9

EJTIR 21(4), 2021, pp.81-103 100
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
Bryan, M. L., & Jenkins, S. P. (2016). Multilevel Modelling of Country Effects: A Cautionary Tale.
European Sociological Review, 32(1), 3–22. https://doi.org/10.1093/esr/jcv059
Camara, F., Romano, R., Markkula, G., Madigan, R., Merat, N., & Fox, C. (2018). Empirical game theory
of pedestrian interaction for autonomous vehicles. In Proceedings of Measuring Behavior 2018.
Manchester: Metropolitan University.
Choi, J. K., & Ji, Y. G. (2015). Investigating the Importance of Trust on Adopting an Autonomous
Vehicle. International Journal of Human-Computer Interaction, 31(10), 692–702.
https://doi.org/10.1080/10447318.2015.1070549
Couper, M. P. (2017). New Developments in Survey Data Collection. Annual Review of Sociology, 43(1),
121–145. https://doi.org/10.1146/annurev-soc-060116-053613
Cunningham, M. L., & Regan, M. A. (2020). Public Opinion About Automated and Self-Driving
Vehicles. In D. L. Fisher, W. J. Horrey, J. D. Lee, & M. A. Regan (Eds.), Handbook of human factors for
automated, connected, and intelligent vehicles (pp. 95–106). Boca Raton: CRC Press, Taylor & Francis Group.
Cunningham, M. L., Regan, M. A., Horberry, T., Weeratunga, K., & Dixit, V. (2019a). Public opinion
about automated vehicles in Australia: Results from a large-scale national survey. Transportation
Research Part a: Policy and Practice, 129, 1–18. https://doi.org/10.1016/j.tra.2019.08.002
Cunningham, M. L., Regan, M. A., Ledger, S. A., & Bennett, J. M. (2019b). To buy or not to buy?
Predicting willingness to pay for automated vehicles based on public opinion. Transportation Research
Part F: Traffic Psychology and Behaviour, 65, 418–438. https://doi.org/10.1016/j.trf.2019.08.012
Davis, F. D., Bagozzi, R. P., & Warshaw, P. R. (1989). User Acceptance of Computer Technology: A
Comparison of Two Theoretical Models. Management Science, 35(8), 982–1003.
https://doi.org/10.1287/mnsc.35.8.982
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. https://doi.org/10.1016/j.trc.2017.03.003
Deb, S., Strawderman, L., Carruth, D. W., DuBien, J., Smith, B., & Garrison, T. M. (2017). Development
and validation of a questionnaire to assess pedestrian receptivity toward fully autonomous vehicles.
Transportation Research Part C: Emerging Technologies, 84, 178–195.
https://doi.org/10.1016/j.trc.2017.08.029
Diekmann, A. (2016). Spieltheorie: Einführung, Beispiele, Experimente (Originalausgabe, 4., überarbeitete
Auflage). rowohlts enzyclopädie. Reinbek bei Hamburg: Rowohlts Enzyklopädie im Rowohlt
Taschenbuch Verlag.
Eriksson, G. (2021). D2.2 Report on the findings of the expert online survey. Retrieved from
http://www.brave-project.eu/wp-content/uploads/2021/03/20210225-Deliverable-D2.2-
Stakeholder-survey-Final.pdf
Fagnant, D. J., & Kockelman, K. (2015). Preparing a nation for autonomous vehicles: Opportunities,
barriers and policy recommendations. Transportation Research Part a: Policy and Practice, 77, 167–181.
https://doi.org/10.1016/j.tra.2015.04.003
GESIS Leibniz Institute for the Social Science (2019). International Social Survey Programme ISSP 2017 –
Social Networks and Social Resources. Variable Report (No. 2019/13). Köln.
Ghazizadeh, M., Lee, J. D., & Boyle, L. N. (2012). Extending the Technology Acceptance Model to assess
automation. Cognition, Technology & Work, 14(1), 39–49. https://doi.org/10.1007/s10111-011-0194-3
Ghazizadeh, M., Peng, Y., Lee, J. D., & Boyle, L. N. (2012). Augmenting the Technology Acceptance
Model with Trust: Commercial Drivers’ Attitudes towards Monitoring and Feedback. Proceedings of the
Human Factors and Ergonomics Society Annual Meeting, 56(1), 2286–2290.
https://doi.org/10.1177/1071181312561481

EJTIR 21(4), 2021, pp.81-103 101
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
Gkartzonikas, C., & Gkritza, K. (2019). What have we learned? A review of stated preference and choice
studies on autonomous vehicles. Transportation Research Part C: Emerging Technologies, 98, 323–337.
https://doi.org/10.1016/j.trc.2018.12.003
Gold, C., Happee, R., & Bengler, K. (2018). Modeling take-over performance in level 3 conditionally
automated vehicles. Accident Analysis & Prevention, 116, 3–13.
https://doi.org/10.1016/j.aap.2017.11.009
Gold, C., Körber, M., Hohenberger, C., Lechner, D., & Bengler, K. (2015). Trust in Automation – Before
and After the Experience of Take-over Scenarios in a Highly Automated Vehicle. Procedia Manufacturing,
3, 3025–3032. https://doi.org/10.1016/j.promfg.2015.07.847
Hair, J. F., Babin, B. J., Anderson, R. E., & Black, W. C. (2014). Multivariate data analysis (7th edition).
Pearson custom library. Harlow, Essex: Pearson.
Hergeth, S., Lorenz, L., Vilimek, R., & Krems, J. F. (2016). Keep Your Scanners Peeled: Gaze Behavior as
a Measure of Automation Trust During Highly Automated Driving. Human Factors: The Journal of the
Human Factors and Ergonomics Society, 58(3), 509–519. https://doi.org/10.1177/0018720815625744
Hoff, K. A., & Bashir, M. (2015). Trust in Automation: Integrating Empirical Evidence on Factors That
Influence Trust. Human Factors: The Journal of the Human Factors and Ergonomics Society, 57(3), 407–434.
https://doi.org/10.1177/0018720814547570
Hohenberger, C., Spörrle, M., & Welpe, I. M. (2016). How and why do men and women differ in their
willingness to use automated cars? The influence of emotions across different age groups. Transportation
Research Part a: Policy and Practice, 94, 374–385. https://doi.org/10.1016/j.tra.2016.09.022
Hudson, J., Orviska, M., & Hunady, J. (2019). People’s attitudes to autonomous vehicles. Transportation
Research Part a: Policy and Practice, 121, 164–176. https://doi.org/10.1016/j.tra.2018.08.018
Hulse, L. M., Xie, H., & Galea, E. R. (2018). Perceptions of autonomous vehicles: Relationships with road
users, risk, gender and age. Safety Science, 102, 1–13. https://doi.org/10.1016/j.ssci.2017.10.001
ISSP Research Group (2019). International Social Survey Programme: Social Networks and Social Resources -
ISSP 2017. https://doi.org/10.4232/1.13322
Jing, P., Xu, G., Chen, Y., Shi, Y., & Zhan, F. (2020). The Determinants behind the Acceptance of
Autonomous Vehicles: A Systematic Review. Sustainability, 12(5), 1719.
https://doi.org/10.3390/su12051719
Johnsen, A., Strand, N., Andersson, J., Patten, C., Kraetsch, C., & Takman, J. (2018). Literature review on
the acceptance and road safety, ethical, legal, social and economic implications of automated vehicles. Deliverable
2.1 from the EU-H2020-project BRAVE – BRidging the gaps for the adoption of Automated VEhicles
(IfeS-Materialien No. 2/2018). Nürnberg.
Kaleefathullah, A. A., Merat, N., Lee, Y. M., Eisma, Y. B., Madigan, R., Garcia, J., & Winter, J. de (2020).
External Human-Machine Interfaces Can Be Misleading: An Examination of Trust Development and
Misuse in a CAVE-Based Pedestrian Simulation Environment. Human Factors, 18720820970751.
https://doi.org/10.1177/0018720820970751
Kaur, K., & Rampersad, G. (2018). Trust in driverless cars: Investigating key factors influencing the
adoption of driverless cars. Journal of Engineering and Technology Management, 48, 87–96.
https://doi.org/10.1016/j.jengtecman.2018.04.006
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, 105441. https://doi.org/10.1016/j.aap.2020.105441
König, M., & Neumayr, L. (2017). Users’ resistance towards radical innovations: The case of the self-
driving car. Transportation Research Part F: Traffic Psychology and Behaviour, 44, 42–52.
https://doi.org/10.1016/j.trf.2016.10.013

EJTIR 21(4), 2021, pp.81-103 102
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
Kraetsch, C., Schrauth, B., Johnsen, A., & Funk, W. (2019). Zwischen Vertrauen und Misstrauen: Die Sicht
von Verkehrsteilnehmern auf automatisierte Fahrzeuge. Paper presented at 3. Kongress der Fachgruppe
Verkehrspsychologie, Saarbrücken.
Kyriakidis, M., Happee, R., & Winter, J. de (2015). Public opinion on automated driving: Results of an
international questionnaire among 5000 respondents. Transportation Research Part F: Traffic Psychology
and Behaviour, 32, 127–140. https://doi.org/10.1016/j.trf.2015.04.014
Lee, J. D., & See, K. A. (2004). Trust in Automation: Designing for Appropriate Reliance. Human Factors:
The Journal of the Human Factors and Ergonomics Society, 46(1), 50–80.
https://doi.org/10.1518/hfes.46.1.50_30392
Liu, P., Du, Y., Wang, L., & Da Young, J. (2020). Ready to bully automated vehicles on public roads?
Accident Analysis & Prevention, 137, 105457. https://doi.org/10.1016/j.aap.2020.105457
Madigan, R., Louw, T., Wilbrink, M., Schieben, A., & Merat, N. (2017). What influences the decision to
use automated public transport? Using UTAUT to understand public acceptance of automated road
transport systems. Transportation Research Part F: Traffic Psychology and Behaviour, 50, 55–64.
https://doi.org/10.1016/j.trf.2017.07.007
Merat, N., Louw, T., Madigan, R., Wilbrink, M., & Schieben, A. (2018). What externally presented
information do VRUs require when interacting with fully Automated Road Transport Systems in shared
space? Accident Analysis & Prevention, 118, 244–252. https://doi.org/10.1016/j.aap.2018.03.018
Milakis, D., van Arem, B., & van Wee, B. (2017). Policy and society related implications of automated
driving: A review of literature and directions for future research. Journal of Intelligent Transportation
Systems, 21(4), 324–348. https://doi.org/10.1080/15472450.2017.1291351
Millard-Ball, A. (2018). Pedestrians, Autonomous Vehicles, and Cities. Journal of Planning Education and
Research, 38(1), 6–12. https://doi.org/10.1177/0739456X16675674
Nordhoff, S., Kyriakidis, M., van Arem, B., & Happee, R. (2019). A multi-level model on automated
vehicle acceptance (MAVA): a review-based study. Theoretical Issues in Ergonomics Science, 20(6), 682–
710. https://doi.org/10.1080/1463922X.2019.1621406
Nordhoff, S., Louw, T., Innamaa, S., Lehtonen, E., Beuster, A., Torrao, G., . . . Merat, N. (2020). Using the
UTAUT2 model to explain public acceptance of conditionally automated (L3) cars: A questionnaire
study among 9,118 car drivers from eight European countries. Transportation Research Part F: Traffic
Psychology and Behaviour, 74, 280–297. https://doi.org/10.1016/j.trf.2020.07.015
Payre, W., Cestac, J., & Delhomme, P. (2014). Intention to use a fully automated car: Attitudes and a
priori acceptability. Transportation Research Part F: Traffic Psychology and Behaviour, 27, 252–263.
https://doi.org/10.1016/j.trf.2014.04.009
Payre, W., Cestac, J., & Delhomme, P. (2016). Fully Automated Driving: Impact of Trust and Practice on
Manual Control Recovery. Human Factors: The Journal of the Human Factors and Ergonomics Society, 58(2),
229–241. https://doi.org/10.1177/0018720815612319
Penmetsa, P., Adanu, E. K., Wood, D., Wang, T., & Jones, S. L. (2019). Perceptions and expectations of
autonomous vehicles – A snapshot of vulnerable road user opinion. Technological Forecasting and Social
Change, 143, 9–13. https://doi.org/10.1016/j.techfore.2019.02.010
Pettigrew, S., Worrall, C., Talati, Z., Fritschi, L., & Norman, R. (2019). Dimensions of attitudes to
autonomous vehicles. Urban, Planning and Transport Research, 7(1), 19–33.
https://doi.org/10.1080/21650020.2019.1604155
Piao, J., McDonald, M., Hounsell, N., Graindorge, M., Graindorge, T., & Malhene, N. (2016). Public
Views towards Implementation of Automated Vehicles in Urban Areas. Transportation Research Procedia,
14, 2168–2177. https://doi.org/10.1016/j.trpro.2016.05.232
Pyrialakou, V., Gkartzonikas, C., Gatlin, J., & Gkritza, K. (2020). Perceptions of safety on a shared road:
Driving, cycling, or walking near an autonomous vehicle. Journal of Safety Research, 72, 249–258.
https://doi.org/10.1016/j.jsr.2019.12.017

EJTIR 21(4), 2021, pp.81-103 103
Schrauth, Funk, Maier and Kraetsch
The acceptance of conditionally automated cars from the perspective of different road user groups
Rouchitsas, A., & Alm, H. (2019). External Human-Machine Interfaces for Autonomous Vehicle-to-
Pedestrian Communication: A Review of Empirical Work. Frontiers in Psychology, 10, 2757.
https://doi.org/10.3389/fpsyg.2019.02757
SAE International (2018). Taxonomy and Definitions for Terms Related to Driving Automation Systems for
On-Road Motor Vehicles. (SAE J3016_201806).
Saleh, K., Hossny, M., & Nahavandi, S. (2017). Towards trusted autonomous vehicles from vulnerable
road users perspective. In 2017 Annual IEEE International Systems Conference (SysCon) (pp. 1–7).
Montreal, QC, Canada: IEEE. https://doi.org/10.1109/SYSCON.2017.7934782
Schieben, A., Wilbrink, M., Kettwich, C., Madigan, R., Louw, T., & Merat, N. (2019). Designing the
interaction of automated vehicles with other traffic participants: design considerations based on human
needs and expectations. Cognition, Technology & Work, 21(1), 69–85. https://doi.org/10.1007/s10111-
018-0521-z
Schoettle, B., & Sivak, M. (2014a). A Survey of Public Opinion about Autonomous and Self-Driving Vehicles
in the U.S., the U.K., and Australia. Michigan. Retrieved from https://deepblue.lib.umich.edu/bitstream-
/handle/2027.42/108384/103024.pdf?sequence=1&isAllowed=y
Schoettle, B., & Sivak, M. (2014b). Public Opinion about Self-Driving Vehicles in China, India, Japan, the U.S.,
the U.K., and Australia. Michigan. Retrieved from https://deepblue.lib.umich.edu/bitstream-
/handle/2027.42/109433/103139.pdf?sequence=1&isAllowed=y
Scholtz, J. (2003). Theory and evaluation of human robot interactions. In Proceedings of the 36th Hawaii
International Conference on System Sciences. IEEE. https://doi.org/10.1109/HICSS.2003.1174284
Schrauth, B., Maier, S., Kraetsch, C., & Funk, W. (2020). Report on the findings of the BRAVE population
survey: Deliverable 2.3 from the EU-H2020-project BRAVE – BRidging the gaps for the adoption of Automated
VEhicles (IfeS-Materialien No. 2/2020). Nürnberg.
Schrauth, B., Maier, S., Kraetsch, C., & Funk, W. (2021). BRAVE population survey [Dataset].
https://doi.org/10.5281/zenodo.4305868
Stanciu, S. C., Eby, D. W., Molnar, L. J., St. Louis, R. M., Zanier, N., & Kostyniuk, L. P. (2018).
Pedestrians/bicyclists and Autonomous Vehicles: How Will They Communicate? Transportation
Research Record: Journal of the Transportation Research Board, 2672(22), 58–66.
https://doi.org/10.1177/0361198118777091
Straub, E. R., & Schaefer, K. E. (2019). It takes two to Tango: Automated vehicles and human beings do
the dance of driving – Four social considerations for policy. Transportation Research Part a: Policy and
Practice. Advance online publication. https://doi.org/10.1016/j.tra.2018.03.005
United Nations (2019). World Population Prospects 2019: File POP/7-1: Total population (both sexes
combined) by five-year age group, region, subregion and country, 1950-2100 (thousands) - Estimates,
1950 - 2020. Retrieved from https://population.un.org/wpp/Download/Files/1_Indicators-
%20(Standard)/EXCEL_FILES/1_Population/WPP2019_POP_F07_1_POPULATION_BY_AGE_BOT
H_SEXES.xlsx
Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User Acceptance of Information
Technology: Toward a Unified View. MIS Quarterly, 27(3), 425. https://doi.org/10.2307/30036540
Wintersberger, S., Azmat, M., & Kummer, S. (2019). Are We Ready to Ride Autonomous Vehicles? A
Pilot Study on Austrian Consumers’ Perspective. Logistics, 3(4), 20.
https://doi.org/10.3390/logistics3040020
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. https://doi.org/10.1016/j.trc.2018.11.018
Zmud, J. P., & Sener, I. N. (2017). Towards an Understanding of the Travel Behavior Impact of
Autonomous Vehicles. Transportation Research Procedia, 25, 2500–2519.
https://doi.org/10.1016/j.trpro.2017.05.281