Investigating the factors influencing the acceptance of fully autonomous cars

Benleulmi, Ahmed Ziad; Blecker, Thorsten Conference Paper Investigating the factors influencing the acceptance of fully autonomous cars Provided in Cooperation with: Hamburg University of Technology (TUHH), Institute of Business Logistics and General Management Suggested Citation: Benleulmi, Ahmed Ziad; Blecker, Thorsten (2017) : Investigating the factors influencing the acceptance of fully autonomous cars, In: Kersten, Wolfgang Blecker, Thorsten Ringle, Christian M. (Ed.): Digitalization in Supply Chain Management and Logistics: Smart and Digital Solutions for an Industry 4.0 Environment. Proceedings of the Hamburg International Conference of Logistics (HICL), Vol. 23, ISBN 978-3-7450-4328-0, epubli GmbH, Berlin, pp. 99-115, https://doi.org/10.15480/882.1449 This Version is available at: https://hdl.handle.net/10419/209304 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-sa/4.0/ Published in: Digitalization in Supply Chain Management and Logistics Wolfgang Kersten, Thorsten Blecker and Christian M. Ringle (Eds.) ISBN 9783745043280, Oktober 2017, epubli Ahmed Ziad Benleulmi, Thorsten Blecker Investigating the Factors Influencing the Acceptance of Fully Autonomous Cars Proceedings of the Hamburg International Conference of Logistics (HICL) –23 CC-BY-SA 4.0 Investigating the Factors Influencing the Acceptance of Fully Autonomous Cars Ahmed Ziad Benleulmi1, Thorsten Blecker1 1 – Hamburg University of Technology Once thought of as a product of science fiction, self-driving cars are discussed today as an unavoidable means towards improving transportation systems. In fact, many car manufacturers have announced their plans to deploy highly autonomous cars as soon as 2020; according to the Society of Automotive Engineers (SAE) these vehicles are capable of reacting “even if the human driver does not respond appropriately to the request to intervene (SAE level 4)”. There is however a long way to go before fully autonomous cars (SAE level 5) - where pedals and steering wheels are forgone and limitations to driving during severe weather or in unmapped areas are surmounted - are produced. Herein, the overall aim is to study the drivers and inhibitors of autonomous cars’ acceptance across cultures with a special focus on the different risks that might deter consumers from using highly and/or fully autonomous cars. After an extensive reviewing of previous works, a research model based on UTAUT2 was developed and accordingly an online survey was conducted in the US and in Germany; 313 valid answers were collected and analyzed. The findings presented here have serious implications both on the academic field as well as the industry, especially in regards to the roles that risks, culture and gender play in the acceptance of fully autonomous cars. Keywords: Fully Autonomous Cars; Technology Acceptance Models; PLS SEM 99 Investigating the Factors Influencing the Acceptance of Fully Autonomous Cars 1 Introduction According to the World Health Organization (WHO), car accidents rank within the top causes of death of people worldwide and rank first if we only consider young people aged 15 to 29 (WHO, 2015); In 94% of the cases, the fault did not lie with the vehicle but with the human drivers, be it drunkenness, drowsiness or distraction (NHTSA, 2015); Sadly, even though the price paid every year is too high, the number of casualties is expected to rise as a result to more people embracing car ownership (Lipson and Kurman, 2016). Supporters of a rapid adoption and quick mainstreaming of autonomous cars believe that their benefits far outweigh their disadvantages; these beliefs are far from baseless as many facts seem to support them; for instance, according to the Eno Center for Transportation (Eno), if 90% of the driven cars in the USA were autonomous “the number of driving related deaths would fall from 32400 per year to 11300“ (Eno, 2013); Other major benefits of autonomous cars are their convenience, ease of use and the freedom they offer to consumers, especially to the old, disabled and those incapable of driving. However, benefits of autonomous cars notwithstanding, they are far from being perfect or at the very least they present many issues that make it hard for consumers to accept them wholeheartedly. In fact many are reluctant to use them for a multitude of reasons; some do not trust them to be safe, secure or private, others avoid them because of social pressure and some consumers have trouble accepting them simply because they love driving. In order to have a deep understanding of the drivers and inhibitors governing the consumers‘ attitudes towards autonomous cars‘ acceptance, we investigated the main probable drivers; the proposed research model comprises constructs from UTAUT2 as well as other relevant constructs that are rooted in the literature and strongly relevant to the context of this research. 2 Theoretical Background and Hypotheses Technology acceptance models have been around since the early days of information system research, their aim –unchanging over the yearsis to investigate the factors influencing the adoption of a technology or its rejection; Several such models emerged over time: TRA (1975), TPB (1985), TAM (1989) and UTAUT (2003) 100 2 Theoretical Background and Hypotheses to mention some of them (Venkatesh et al., 2003). UTAUT2, an extended version of UTAUT proposed by Venkatesh in 2012, presents a fitting basis for our present research; some of its constructs, pertaining to usefulness (Performance Expectancy), ease of use (Effort Expectancy), social pressure (Social influence) and enjoyment (Hedonic Motivation), should play a major role considering the nature of the present research (Venkatesh, Thong and Xu, 2012). Certainly, the next step is to adapt the theory to the current context; Hong et al (2014) clearly defined the approaches for the contextualization of a theory (Hong et al., 2014); the first level of this process is to add or remove core constructs; Next, contextual factors such as antecedentsare incorporated in the model. Following these guidelines we assimilated some key constructs –rooted in the literatureinto the model (see figure 1); these constructs are Desirability of Control (DEC) defined as “the fear of losing control over the vehicle” (Planing, 2014), Perceived Convenience (PC) which is “the level of convenience toward time, place and execution that one feels when driving an autonomous car” (Hsu and Chang, 2013), Personal Innovativeness in IT (PIIT) defined as an “individual trait reflecting a willingness to try out any new technology” (Agarwal and Prasad, 1998) and the Intention to Prefer an autonomous car over a conventional car (IP). The hypotheses regarding the influence of the previously mentioned constructs on the Behavioral Intention (BI), i.e. the intention of the consumer to use fully autonomous cars, are the following: H1: PIIT has a positive influence on BI H1: DEC has a negative influence on BI H3: HM has a positive influence on BI H4: PE has a positive influence on BI H5: EE has a positive influence on BI H6: SI has a positive influence on BI H7: PC has a positive influence on BI H8: BI has a positive influence on IP Additionally, risks are expected to play a major role inhibiting the acceptance of autonomous cars. we singled out five relevant types of risks, these risks are (1) Privacy Risk (PRIV) linked to a “possible loss of privacy as a result of a voluntary or surreptitious information disclosure to the autonomous car” (Dinev and Hart, 2006; Liao, Liu and Chen, 2011), (2) Performance Risk (PERR) associated with “The 101 Investigating the Factors Influencing the Acceptance of Fully Autonomous Cars Figure 1: Proposed research model 102 2 Theoretical Background and Hypotheses possibility of the autonomous car malfunctioning and not performing as it was designed and advertised and therefore failing to deliver the desired benefits.” (Grewal, Gotlieb and Marmorstein, 1994), (3) Safety Risk (SAFE) which is the risk of the user’s safety being endangered through his use of an autonomous car, (4) Financial Risk (FINR) pertaining to “the potential monetary outlay associated with the initial purchase price as well as the subsequent maintenance cost of autonomous cars” and finally (5) Socio-psychological Risk (SPR) defined as the “Potential loss of status in one’s social group as a result of using fully autonomous cars, looking foolish or untrendy and risking to lower the consumer’s self image” (Kim, Lee and Jung, 2005). Understanding the influence risks have on trust is a major point in this research as it will not only show which risks are relevant but also which ones have more impact. Trust (TRUST) is defined by Mayer et al. (2011) as “the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the truster, irrespective of the ability to monitor or control that other party”. We believe the disposition of a person to trust might also play a role in him/her trusting autonomous cars, hence the DTRUST construct, defined as “how a person sees himself/herself in regards to his/her interactions with other people”(Srivastava, Singh and Srivastava, 2013). The hypotheses pertaining to trust are the following: H9: PERR has a negative influence on TRUST H10: FINR has a negative influence on TRUST H11: PRIV has a negative influence on TRUST H12: SAFE has a negative influence on TRUST H13: SPR has negative influence on TRUST H14: DTRUST has a positive influence on TRUST H15: TRUST has a positive influence on BI 103 Investigating the Factors Influencing the Acceptance of Fully Autonomous Cars 3 Methodology 3.1 Sample Description An online study was conducted in Germany and the USA to collect the necessary data for the present study. The questionnaire was designed following the wellestablished principles for survey design by Dillman, Tortora and Bowker (1998). The survey took approximately 10 minutes to complete and was accessible for three weeks starting from May 31st, 2017. 313 participants answered the survey of which 160 reside in the USA and 153 in Germany. The survey was designed in a way that participants had to answer all questions before they were able to submit the questionnaire. Age distribution shows the mean age of the respondents to be 35; it also shows 41.9% of them to be male and 58.1% to be female (see table 1). The majority of the participants have an annual income inferior to € 50.000. 95.5% of the respondents have had some experience driving cars and 84.3% of them currently own one. 104 3 Methodology Table 1: Participant demographics Variable Category Freq In% Mode Gender Male 131 41.9 Female Female 182 58.1 Age Younger than 35 175 55,9 29 35 and older 138 44.1 Education 8th grade or less 2 0.6 Graduated from college, graduate or post-graduate school Some high school (Grade 9-11) 14 4.5 Graduated from high school 81 25.9 1-3 years of college/university 74 23.6 Graduate or postgraduate 138 44.1 No answer 1 0.3 Annual Income Less than €50.000 157 50.2 Less than €50.000 €50.000 to €100.000 88 28.1 €100.000 to €150.000 14 4.5 €150.000 and more 3 1 I would rather not say 51 16.3 Car Ownership Yes 264 84.3 Yes No 49 15.7 Driving Experience Yes 299 95.5 Yes No 14 4.5 105 Investigating the Factors Influencing the Acceptance of Fully Autonomous Cars Table 4: Group moderation - results of the parametric test Path/Moderator Gender Age Country BI ->IP 2.03** 0.78 0.29 DEC ->BI 0.63 1.39 0.29 DTRUST ->TRUST 2.24** 0.99 0.70 EE ->BI 1.06 1.14 0.10 FINR ->TRUST 1.65 0.18 0.49 HM ->BI 1.94* 0.27 0.82 PC ->BI 0.33 1.02 0.00 PE ->BI 0.53 0.82 0.33 PERR ->TRUST 0.55 0.96 0.47 PIIT ->BI 0.59 0.07 0.34 PRIV ->TRUST 0.4 0.29 1.36 SAFE ->TRUST 0.34 0.06 0.74 SI ->BI 2.24** 0.12 0.81 SPR ->TRUST 1.22 1.55 2.59*** TRUST ->BI 0.49 0.53 0.25 Significance level: *p<.10 **p<.05 ***p<.01 Confidence level: *90% **95% ***99% 112 5 Discussion and Conclusion a moderator, findings show that socio-psychological risks are strong in Germany contrary to the USA where they were found to be insignificant. It is worth noting that in regards to effect sizes (f 2 values), a strong effect size was recorded for BI > IP (1.79); SI > BI (0.17) had the only registered medium effect and all the remaining effects were small, with the exception of FINR > TRUST, PC > BI, PE > BI and PRIV > TRUST (the rejected hypotheses) where the recorded effect sizes were inferior to 0.02 and therefore deemed insignificant. 5 Discussion and Conclusion The main aim of this study was to identify the drivers and inhibitors governing the consumer‘s behavior in regards to the acceptance of fully autonomous cars. For that purpose, we conducted an online survey in two countries namely Germany and the USA; the collected data was then analyzed using PLS SEM. The results of the analysis shed light on the main drivers influencing consumers’ intention to use fully autonomous cars; they also allowed to gauge the role these drivers play in terms of impact. Our findings showed that many factors positively influence the user’s intention to use an autonomous car; and while some of these factors are related to the technology itself such as it being “easily used” or “enjoyable”, others are associated with the user’s own personality such as “personal innovativeness” or the influence of the environment on him/her; The key inhibitors to autonomous cars’ acceptance were found to be the consumers’ thirst for control and risks; Some risks were recorded to have a great influence on a consumer trusting a car, these risks are: the risk of the car being unsafe and the risk of the car malfunctioning. Surprisingly, the performance expectancy of the autonomous cars as well as their convenience were found not to be significant, this can be explained as consumers being more influenced by the negative aspect i.e. the car malfunctioning as well as the unavailability of the car in the market at the present time. Risks that were ascertained to be extraneous are the financial risk as well as the privacy risk as many people did not find them to be “deal breakers”. In terms of implications for theory, the contributed research model comprises all the key elements that play a major role in the acceptance or rejection of fully autonomous cars; as such it can be used in future research as a reference for a contextualized model of technology acceptance in the automobile industry. In terms 113 Investigating the Factors Influencing the Acceptance of Fully Autonomous Cars of managerial and practical implications, the current study presented findings that clarified which elements have an impact -and to which levelmaking it easier for stakeholders to make a better advised decision when dealing the inhibitors to the acceptance of autonomous cars or when promoting the attractive qualities of thesevehicles. Advertizingthebenefitsof the carshouldhelp people havea better understanding of what to expect in terms of productivity and comfort. Decision makers should find ways to promote trust in fully autonomous vehicles, and that through dealing with the consumer concerns’ linked to the performance of the car and to the safety of the user, as these were found to be the strongest inhibiting risk factors; they should also find ways to soften and implicate the people who prefer to have constant control while commuting. Findings related to gender, culture and personal innovativeness should ultimately help devise improved strategies to target these specific population segments in a more effective way. References Agarwal, R. and J. Prasad (1998). “A Conceptual and Operational Definition of Personal Innovativeness in the Domain of Information Technology”. In: Info. Sys. Research 9.2, pp. 204–215. Anderson, J. C. and D. W. Gerbing (1988). “Structural equation modeling in practice: A review and recommended two-step approach.” In: Psychological bulletin 103.3, p. 411. Bagozzi, R. P. and Y. Yi (1988). “On the evaluation of structural equation models”. In: Journal of the academy of marketing science 16.1, pp. 74–94. Chin, W. W. (1998). “The partial least squares approach for structural equation modeling.” In: George A. Marcoulides. Chap. 10, pp. 295–336. Chin, W. W. (2010). “How to Write Up and Report PLS Analyses”. en. In: Handbook of Partial Least Squares. Ed. by V. E. Vinzi, W. W. Chin, J. Henseler, and H. Wang. Springer Handbooks of Computational Statistics. Springer Berlin Heidelberg, pp. 655–690. Dillman, D. A., R. D. Tortora, and D. Bowker (1998). “Principles for constructing web surveys”. In: Joint Meetings of the American Statistical Association. Dinev, T. and P. Hart (2006). “An Extended Privacy Calculus Model for E-Commerce Transactions”. en. In: Information Systems Research 17.1, pp. 61–80. Eno (2013). Preparing a Nation for Autonomous Vehicles. Tech. rep. Grewal, D., J. Gotlieb, and H. Marmorstein (1994). “The Moderating Effects of Message Framing and Source Credibility on the Price-Perceived Risk Relationship”. en. In: Journal of Consumer Research 21.1, p. 145. Hair, J. F., C. M. Ringle, and M. Sarstedt (2011). “PLS-SEM: Indeed a Silver Bullet”. In: The Journal of Marketing Theory and Practice 19.2, pp. 139–152. Hair, J. F., G. T. M. Hult, C. Ringle, and M. Sarstedt (2014). A primer on partial least squares structural equations modeling (PLS-SEM). English. Los Angeles: SAGE. 114 References Hair, J. F., M. Sarstedt, L. Hopkins, and V. G. Kuppelwieser (2014). “Partial least squares structural equationmodeling(PLS-SEM):Anemergingtoolinbusinessresearch”.en.In:EuropeanBusiness Review 26.2, pp. 106–121. Henseler, J., C. M. Ringle, and R. R. Sinkovics (2009). “Advances in International Marketing”. en. In: Advances in International Marketing 20. Ed. by J. Henseler, C. M. Ringle, and R. R. Sinkovics. Hong, W., F. K. Y. Chan, J. Y. L. Thong, L. C. Chasalow, and G. Dhillon (2014). “A Framework and GuidelinesforContext-SpecificTheorizingin InformationSystems Research”.en.In:Information Systems Research 25.1, pp. 111–136. Hsu, H. - h. and Y. - y. Chang (2013). “Extended TAM Model: Impacts of Convenience on Acceptance and Use of Moodle.” In: Online Submission 3.4, pp. 211–218. Hulland, J. (1999). “Use of partial least squares (PLS) in strategic management research: a review of four recent studies”. en. In: Strategic Management Journal 20.2, pp. 195–204. Kim, H. K., M. Lee, and M. Jung (2005). “Perceived Risk and Risk-Reduction Strategies For High- Technology Services”. In: 6, pp. 171–179. Liao, C., C. - C. Liu, and K. Chen (2011). “Examining the impact of privacy, trust and risk perceptions beyond monetary transactions: An integrated model”. en. In: Electronic Commerce Research and Applications 10.6, pp. 702–715. Lipson, H. and M. Kurman (2016). Driverless: intelligent cars and the road ahead. Cambridge, Massachusetts: The MIT Press. Mayer, P., D. Volland, F. Thiesse, and E. Fleisch (2011). “User Acceptance of’Smart Products’: An Empirical Investigation”. In: NHTSA (2015). Critical Reasons for Crashes Investigated in the National Motor Vehicle Crash Causation Survey.pdf. Tech. rep. NHTSA’s National Center for Statistics and Analysis. Planing,P.(2014).InnovationAcceptance.en.DOI:10.1007/978-3-658-05005-4.Wiesbaden:Springer Fachmedien Wiesbaden. Reinartz, W., M. Haenlein, and J. Henseler (2009). “An empirical comparison of the efficacy of covariance-based and variance-based SEM”. en. In: International Journal of Research in Marketing 26.4, pp. 332–344. Srivastava, R. K., S. Singh, and V. Srivastava (2013). “The Role of Trust and Technology Acceptance Model(TAM)onCustomerAcceptanceof Mobile Banking: An IntegratedModel toImproveMobile Banking in Banking Sector-A Conceptual Paper”. en. In: International Journal of Asian Business and Information Management 4.1, pp. 31–43. Venkatesh, V., M. G. Morris, G. B. Davis, and F. D. Davis (2003). “User acceptance of information technology: Toward a unified view”. In: MIS quarterly, pp. 425–478. Venkatesh, V., J. Y. Thong, and X. Xu (2012). “Consumer acceptance and use of information technology: extending the unified theory of acceptance and use of technology”. In: MIS quarterly 36.1, pp. 157–178. Vinzi, V. E., L. Trinchera, and S. Amato (2010). “PLS Path Modeling: From Foundations to Recent Developments and Open Issues for Model Assessment and Improvement”. en. In: Handbook of Partial Least Squares. Ed. by V. E. Vinzi, W. W. Chin, J. Henseler, and H. Wang. Springer Handbooks of Computational Statistics. Springer Berlin Heidelberg, pp. 47–82. WHO (2015). Global status report on road safety 2015. English. OCLC: 985423312. 115