A Model of EV Adoption and Rank-Based Contributing Factors

Term Start:

June 1, 2024

Term End:

May 31, 2025

Budget:

$150,000

Keywords:

Electric Vehicles (EVs), EV Adoption Behavior, EV Ownership, Survey Datasets

Thrust Area(s):

Data Modeling and Analytic Tools, Equity and Understanding User Needs

University Lead:

The University of Texas at Austin

Researcher(s):

Chandra Bhat

The transportation sector is one of the largest contributors to greenhouse gas emissions and accounts for nearly 45% of global oil demand, primarily due to the widespread use of internal combustion engine vehicles (ICEVs). In fact, while the United States has seen overall reductions of greenhouse gas emissions of around 3% since 1990, emissions in the transportation sector have increased by over 20% in the same period. Electric Vehicles (EVs) potentially serve as an important “vehicle” to combat climate change and reduce transport-related greenhouse gas emissions. However, while EV adoption rates have been growing over the past several years, EV costs remain higher than ICEV costs, adoption rates have varied dramatically across different segments of the population, and the share of EVs on the road remains relatively low.

Because EV market shares have so far been very limited, few studies have modeled EV adoption at an individual level using actual EV owners. Instead, much of the literature is focused on two areas: (1) studies of the intention to buy EVs using stated preference surveys, and (2) examinations of aggregate trends in EV ownership levels. But some comparisons between these methods suggest that there is a disconnect between adoption intentions and observed aggregate EV ownership levels. Thus, there is a need for a detailed individual-level examination of EV adoption behavior using the revealed preferences of actual EV owners. Additionally, many studies focus broadly on the issue of EV adoption without considering the differences between major types of EVs, specifically plug-in hybrid electric vehicles (PHEVs) and battery electric vehicles (BEVs). Given that different characteristics of these EV types may appeal to different users, there is a need to understand adoption patterns by EV type.

In the above context, we examine EV adoption behavior using a survey of California households that includes revealed adoption behaviors as well as a ranked set of factors that led to adoption for existing EV owners. The survey was undertaken between November 2022 and January 2023 and collected a total of 1,311 responses from a segment of EV and solar panel owners and a second representative sample of California homeowners. Using the survey, we employ a Generalized Heterogeneous Data Model to jointly model adoption for BEVs and PHEVs along with a rank-based model of the factors that contribute to EV adoption among owners. The use of a joint modeling approach that accommodates unobserved correlations between these outcomes allows for the possibility of “self-selection” effects among EV owners (that is, EV owners may be more likely to prioritize specific reasons for choosing EVs) and allows us to extend the results (of EV adoption and reasons for any such adoption) to the entire population of current EV and non-EV owners, well beyond the population of existing EV owners. We also incorporate four latent constructs representing lifestyle preferences and perceptions of EVs, each of which has been shown to play a role in EV adoption. The results of this study will have important implications for social and environmental policies promoting EV adoption and will also inform transportation planners in the design of EV charging infrastructure.

Project scope

Scroll to Top