Assessing public opinions of and interest in new vehicle technologies: An Austin perspective

Car travel is relatively unsafe, costly, and burdensome. Connected and autonomous vehicles (CAVs) offer a potential solution, promising to significantly reduce the high proportion of crashes attributed to driver error. CAVs represent a major technological leap in personal transportation, with the potential for safer and more convenient travel. Several mainstream companies are already developing and testing prototypes, and various jurisdictions are enacting legislation to allow testing on public roads. Predictions suggest a significant market share for autonomous-capable vehicles in the coming decades. However, successful implementation hinges on public acceptance and adoption, requiring an understanding of public perception and willingness to pay (WTP) for these technologies. Existing studies provide descriptive statistics, but lack econometric analysis crucial for policymaking. This study addresses this gap by surveying Austin, Texas residents to analyze their opinions and WTP for CAVs, considering various factors like demographics, built environment, and travel characteristics. The study also examines SAV adoption rates under different pricing scenarios, the timing of AV adoption influenced by social pressure, and potential residential shifts due to changes in travel patterns. The behavioral models developed in this research will be useful in forecasting long-term CAV technology adoption under various scenarios.

This section reviews recent public opinion surveys on CAV adoption. Studies like Casley et al. (2013), Begg (2014), Kyriakidis et al. (2014), and Schoettle and Sivak (2014a, 2014b) reveal varying levels of public awareness and interest in CAVs. These studies highlight safety as a major driver of adoption, along with concerns about equipment failure, legal liability, and hacking. Willingness to pay varies considerably across studies and depends on factors like automation level, insurance incentives, and demographic characteristics. The studies also show differing opinions regarding the likely benefits of CAVs, such as fuel economy, travel time savings, and reduced congestion. Underwood (2014) surveyed experts, who identified legal liability as a major barrier to CAV deployment. Other surveys from CarInsurance.com (Vallet, 2013), J.D. Power (2012), KPMG (2013), and Continental (2015) provide additional insights into public perception, emphasizing the role of incentives and age-related differences in WTP. However, these prior studies lack a comprehensive econometric analysis that integrates various factors and offers a predictive model for long-term adoption. This study aims to bridge this gap by providing such an econometric analysis based on a survey of Austinites.

Data was collected via an online survey administered in Austin, Texas from October to December 2014 using Qualtrics. The survey included questions on respondents' perceptions of AV technology, ridesharing, carsharing, WTP for CAVs, SAV adoption rates, home location decisions, adoption timing of AVs, travel patterns, and demographics. The initial sample of 510 respondents was refined to 347 Austinites after excluding non-residents and scaling for demographic bias using the 2013 American Community Survey. Respondents' home addresses were geocoded to incorporate built-environment factors from Austin's traffic analysis zones (TAZs). The study employed univariate and bivariate ordered probit (OP) models to analyze the response variables, including WTP for Level 3 and Level 4 AVs, SAV adoption rates under different pricing scenarios, adoption timing of AVs, and home location decisions. Model specifications included demographic, built-environment, zone-level, and technology-related variables. Stepwise elimination was used to select significant predictors, considering both statistical and practical significance. Practical significance was assessed by evaluating the change in predicted probabilities resulting from a one-standard-deviation change in the explanatory variables.

The survey revealed that Austinites are relatively tech-savvy, with high rates of smartphone ownership and familiarity with ridesharing and carsharing services. The average WTP for Level 4 automation was significantly higher ($7253) than for Level 3 automation ($3300). Equipment failure was identified as the primary concern regarding AVs, while fewer crashes were perceived as the most likely benefit. Approximately 24% of respondents were willing to pay over $5000 for Level 3 automation, and 57% for Level 4 automation. Regarding SAV adoption, higher adoption rates were predicted for full-time male workers living in urban areas, while licensed drivers showed less interest. The findings highlight the significant impact of pricing, with higher prices leading to lower SAV adoption rates. The adoption timing of AVs showed that tech-savvy individuals, particularly males with higher incomes living in urban areas, were more likely to adopt AVs quickly, with less dependence on their friends' adoption rates. Finally, regarding home location shifts, respondents with more children and those living in higher employment density areas were more likely to move farther from the city center, while those living in densely populated areas showed a preference to stay closer to the city center after widespread AV and SAV adoption. In summary, various aspects of willingness to pay, adoption rates, and relocation decisions were explained through ordered probit models, identifying significant relationships between respondents' demographics, built-environment factors, and travel characteristics and their attitudes toward CAVs.

The findings address the research question by providing insights into the factors influencing public opinion and adoption of CAV technologies. The results demonstrate the importance of considering demographic factors, built-environment characteristics, and individual travel patterns when forecasting CAV adoption and developing related policies. The higher WTP for Level 4 AVs compared to Level 3 AVs highlights the potential market for fully autonomous vehicles. The strong influence of technology savviness and income suggests that targeted marketing campaigns and policy incentives could encourage broader adoption. The findings regarding SAV adoption rates underscore the need for careful pricing strategies to maximize market penetration. The observed correlations between home location decisions and CAV adoption highlight the potential for significant changes in urban land use patterns. The study's econometric models provide a valuable tool for policymakers and planners to simulate long-term adoption scenarios and assess the impact of different policy interventions.

This study provides valuable insights into the factors influencing public opinion and adoption of connected and autonomous vehicles (CAVs) in Austin, Texas. Key findings include a significantly higher willingness to pay for Level 4 automation, the importance of technology savviness and income in adoption decisions, and the potential for significant changes in urban land use patterns due to CAV adoption. The study's econometric models offer a useful tool for predicting long-term adoption rates and inform the development of effective policies and strategies to facilitate a smooth transition to CAVs. Future research could explore the generalizability of these findings to other contexts, investigate the impact of specific policy interventions, and further refine the models by incorporating additional variables and data sources.

The study's findings are based on a survey of Austin residents, which may not be fully generalizable to other regions or populations. The sample, while scaled to address demographic bias, might not perfectly represent the entire Austin population. The reliance on stated preferences, rather than revealed preferences, could lead to a potential gap between stated intentions and actual behavior. Furthermore, the study's temporal scope is limited to a specific time period, and evolving technologies and public awareness may affect the results over time.