Examining the Influence of Technological Perception, Cost, and Accessibility on Electric Vehicle Consumer Behavior in Thailand

Abstract

This study investigates consumer behavior in electric vehicle (EV) adoption, focusing on how factors like convenience, accessibility, technological perception, and cost influence the travel patterns and usage behavior of EV drivers in Thailand. This study aims to address the research gap in the comparative behavior between electric vehicles and public transport in a developing country. Using a quantitative approach, the study collected data via surveys distributed online and face-to-face interviews with a stratified sample of 398 respondents. The survey assessed the relationships between convenience and accessibility, technology perception, cost of ownership, and travel patterns using structural equation modeling (SEM). The findings reveal that convenience and accessibility significantly affect consumer perceptions of technology and the cost of ownership, which, in turn, influences travel patterns. Technology perception and performance serve as partial mediators, suggesting that improving the infrastructure enhances EV adoption. Additionally, the cost of ownership, including long-term savings, positively impacts usage behavior. This study provides key insights for policymakers and urban planners aiming to promote the adoption of EVs. Enhancing charging infrastructure, offering government incentives, and improving public awareness of long-term cost benefits are recommended strategies. These findings are particularly relevant in urban environments and offer guidance for developing infrastructure policies that align with consumer preferences.

1. Introduction

The rapid growth of urbanization and environmental awareness in Thailand have led to significant changes in transportation preferences, particularly in the adoption of electric vehicle (EVs) systems. As a country strives to reduce its carbon footprint and enhance sustainable mobility, understanding the factors that influence consumer behavior in choosing EVs systems is crucial. This examination of consumer behavior in electric vehicle adoption aims to explore the underlying motivations, barriers, and socioeconomic factors that drive the adoption of this form of green transportation.

The transportation sector in Thailand is a major contributor to greenhouse gas emissions, accounting for a significant portion of the country’s overall carbon output. The transportation sector in Thailand is a major emitter, accounting for approximately 29% of economy-wide CO₂ emissions by 2023 (~79 Mt CO₂) [1]. In response to this challenge, the Thai government has implemented policies and incentives to promote the adoption of EVs and expand the MRT (Mass Rapid Transit) network, particularly in Bangkok and other major cities. The growing availability of EVs, coupled with advancements in charging infrastructure, has made them increasingly viable options for consumers. Simultaneously, the expansion of MRT lines has provided a reliable and efficient alternative to traditional modes of public transport such as buses and taxis, which are often plagued by traffic congestion.

Consumer behavior in the context of transportation is influenced by a variety of factors including cost, convenience, environmental impact, and social status [2]. In the case of EVs, early adopters are often motivated by environmental concerns, the desire to reduce dependence on fossil fuels, and potential long-term savings on fuel and maintenance costs. However, barriers such as high upfront costs, limited charging infrastructure, and concerns about battery life and range anxiety still pose challenges for widespread adoption. Studies have shown that government incentives such as tax rebates and subsidies play a crucial role in mitigating these barriers and encouraging consumers to switch to EVs [3,4].

EV types and consumer perceptions. Electric vehicles vary by powertrain architecture: battery electric vehicles (BEVs) rely solely on grid-charged batteries, plug-in hybrid electric vehicles (PHEVs) combine a chargeable battery with an internal combustion engine, and hybrid electric vehicles (HEVs) use regenerative charging without external plug-in capability. These differences shape consumers’ perceptions. Charging convenience/accessibility is most salient for BEVs (public/home charging density and charging time), somewhat salient for PHEVs (shorter electric-only range and gasoline fallback), and minimal for HEVs. Range anxiety and trip planning burden are typically the highest for BEVs, lower for PHEVs, and negligible for HEVs. The perceived environmental benefit tends to rank BEV ≥ PHEV ≥ HEV, while the perceived cost of ownership/usage weighs electricity prices, fuel savings, battery longevity, and maintenance differently across types. Accordingly, the salience and direction of belief-based constructs (e.g., perceived usefulness, ease of use, and risk) may vary by EV type, implying the potential moderation of paths in our framework (e.g., charging convenience → perceived usefulness stronger for BEVs than HEVs).

This study aims to conduct a comparative study of consumer behavior between electric vehicles in Thailand. By examining the convenience and accessibility, technology perception and performance, cost of ownership, and usage factors that drive consumer preferences, this study seeks to uncover the underlying motivations and barriers that influence the adoption of electric vehicle transportation options. Additionally, this research will explore how trip purpose, distance traveled, cost, frequency of use, and multimodal travel behavior affect consumers’ decision-making processes.

Research Gap: Insights from Previous Studies

While extensive research has been conducted on consumer behavior related to both electric vehicles and mass rapid transit systems, there remains a significant gap in the comparative analysis of these two transportation modes within the context of a developing country such as Thailand. Much of the existing literature has focused on either EV adoption or public transit use in isolation, often in the context of developed economies with established infrastructure and differing cultural attitudes towards transportation. For example, studies on electric vehicle adoption have largely concentrated on factors such as environmental awareness, economic incentives, and availability of charging infrastructure [5,6]. These studies have provided valuable insights into the challenges and opportunities associated with promoting EVs, particularly in terms of consumer attitudes towards sustainability and perceived costs and benefits of EV ownership. However, these studies often overlook the comparative appeal of public transit systems, especially in urban environments where space is limited and the cost of car ownership is high.

The research gap, therefore, lies in the intersection of these two fields: understanding how consumers choose between electric vehicles in a rapidly urbanizing environment such as Thailand. While some studies have begun to explore this area, they often lack a comprehensive analysis that considers the full range of factors influencing consumer behavior, including the economic, environmental, social, and psychological dimensions. Similarly, research on EV adoption in Thailand, such as that of Pojani and Stead [7], has not sufficiently considered the competitive role of public transit in shaping consumer decisions. As a result, there is a need for research that bridges these two areas and provides a holistic view of the factors that influence consumer choices in the context of Thailand’s evolving transportation landscape.

This study aims to address this research gap by examining consumers’ behavior towards electric vehicles in Thailand. By integrating insights from both fields, this research contributes to a more nuanced understanding of how consumers make transportation decisions in an urban environment where both individual and collective mobility options are increasingly viable. The findings of this study contribute to a deeper understanding of the dynamics between individual and collective transportation choices in urban settings. Moreover, it will provide valuable insights for policymakers, industry stakeholders, and urban planners in designing more effective transportation policies that align with consumer preference goals.

Structure of the paper. The remainder of this article is organized as follows: Section 2 reviews the literature on electric-vehicle (EV) adoption and develops the study hypotheses. Section 3 describes the study’s data, sampling, measures, and analytical procedures. Section 4 presents the empirical results. Section 5 discusses the theoretical contributions as well as managerial and policy implications. Section 6 concludes the paper by highlighting the limitations and directions for future research.

2. Literature Review

The theoretical framework guiding the comparative study of consumer behavior between electric vehicles (EVs) is based on several interdisciplinary theories that seek to explain how individuals make transportation choices. This study leverages concepts from behavioral economics, technology adoption theories, and urban transportation planning to build a comprehensive understanding of the factors influencing consumer behavior.

2.1. Behavioral Economics and Consumer Decision-Making

Behavioral economics provides a foundation for understanding the psychological and economic factors that influence consumer choice. According to this theory, consumers do not always act rationally; instead, their decisions are influenced by cognitive biases, social preferences, and the perceived utility of available options [8]. In the context of transportation, consumers weigh factors such as cost, convenience, and social status; however, their choices are also shaped by habits, perceptions of environmental impact, and government incentives [3].

2.2. Technology Acceptance Model (TAM)

The Technology Acceptance Model [9] is central to understanding how consumers adopt new technologies like electric vehicles. The TAM posits that perceived usefulness and perceived ease of use are the primary factors determining an individual’s intention to use a technology. Applied to EVs, this model helps explain how factors such as technological advancements, charging infrastructure, and concerns regarding range anxiety influence consumer adoption. The model also sheds light on how technological perceptions and performance, such as the reliability and innovation associated with EVs, affect consumer behavior.

2.3. Integration of Theories in the Conceptual Framework

The conceptual framework for this study integrates these theories to explore how cost, convenience, environmental considerations, and social influences shape consumers’ preferences for electric vehicle systems. The study hypothesizes that consumer behavior is not only a function of economic rationality but also of technological acceptance and accessibility. It considers how these factors interact to influence travel patterns, usage behavior, and the overall decision-making process in urban environments. This theoretical foundation supports this study’s goal of filling the research gap identified in the literature. It offers a comprehensive approach to understanding the dynamics between individual (EVs) transportation choices in a rapidly urbanizing context, such as Thailand.

2.4. Research Hypothesizes

2.4.1. Convenience & Accessibility and Technology Perception & Performance

The relationship between convenience, accessibility, technology perception, and performance for electric vehicle (EV) users is shaped by several factors, including the availability of charging infrastructure, ease of use, and overall user satisfaction. In the context of electric vehicles (EVs) and mass rapid transit systems, the relationship between convenience, accessibility, and technological perception and performance is multifaceted in the context of EVs and mass rapid transit systems. Convenience and accessibility often directly influence users’ perceptions of technology, particularly in the adoption of new transportation technologies, such as EVs and mass rapid transit.

The availability of charging infrastructure is a key determinant of the adoption and perception of electric vehicles. Research indicates that both actual accessibility (such as the number of charging stations) and perceived accessibility (drivers’ sense of how accessible these stations are) have a substantial influence on the decision to purchase and use EVs [10]. With WEVC technology moving toward commercialization, evaluating thermal risks in high-power fast charging becomes crucial. This study examines risk factors associated with magnetic couplers and foreign objects (FOs) using an electromagnetic–thermal coupled model. The results quantify power losses and temperature rises and introduce a four-level temperature-based framework for risk classification [11]. The perception of electric vehicles as a cutting-edge technology is a major factor in their adoption. Drivers who view EVs as superior to conventional vehicles, particularly in terms of environmental benefits and long-term cost savings, are more inclined to adopt them. However, concerns related to range, safety, and reliability can negatively affect this perception, thereby influencing the overall acceptance of EVs [12]. Driving style and usage patterns significantly influence the performance of electric vehicles. For example, aggressive driving can result in increased energy consumption and decreased battery lifespan, which can alter the perception of the vehicle’s performance. Additionally, the adoption of EVs often leads to changes in driving behavior, with drivers typically modifying their habits to enhance battery efficiency [13]. The interplay between convenience and accessibility plays a critical role in shaping the technological perception and performance of electric vehicles (EVs) and mass rapid transit systems. Research highlights the significant influence of accessibility on the adoption of EVs, where factors such as objective, perceived, and future accessibility collectively affect individuals’ decisions to purchase EVs [10]. Additionally, the integration of rest and charging behaviors during long-distance travel helps mitigate the limitations posed by EVs’ battery capacity, thereby making the intercity travel experience more comparable to that of traditional fuel-powered vehicles [14]. Perceived accessibility, which encompasses elements such as safety and service quality, is crucial for transportation planning and evaluation. This is driving a shift towards accessibility-focused planning for the development of transportation systems [15]. User perceptions of convenience, such as ease of charging and vehicle performance, significantly influence EV adoption and continued use. Users who find EVs convenient and well suited to their daily routines are more likely to develop positive attitudes toward these vehicles, thereby increasing adoption rates [16]. Access to reliable charging infrastructure plays a key role in enhancing user satisfaction and perceptions of EV technology. For instance, advancements in battery performance and charging infrastructure have been shown to significantly improve overall accessibility, which in turn positively influences users’ views of EVs as a practical transportation option [17]. Convenience and accessibility are critical factors that significantly impact users’ perceptions and perceived performance of EV technology. Improving access to charging infrastructure and ensuring the ease of using EVs can enhance user satisfaction and encourage the wider adoption of electric vehicles. Enhancements in convenience, accessibility, and favorable perceptions of technology play a crucial role in the successful adoption and performance of electric vehicles. As these aspects improve, the probability of broader EV adoption and increased driver satisfaction correspondingly increase. Building on the foregoing literature, we derive testable hypotheses regarding EV purchase intention and its determinants.

As shown in Figure 1, we set out directional, testable propositions connecting CA, TP, and CU to travel patterns and usage (TU). The hypotheses state the expected signs, identify key mediations (e.g., CA → TP → TU), and note possible moderators (EV type; urbanicity). They are derived from established technology adoption and travel behavior literature and serve as the blueprint for the SEM in Section 5, structuring the interpretation of the results.

Hypothesis H1. 

Convenience & accessibility have a positive and significant effect on technology perception & performance.

2.4.2. Convenience & Accessibility and Cost of Ownership & Usage

The total cost of ownership (TCO) for electric vehicles (EVs) can become competitive with that of conventional vehicles, especially as battery costs decline and incentives become more favorable. However, the TCO remains highly dependent on driving habits, vehicle range, and accessibility of the charging infrastructure [18]. The total cost of ownership (TCO) can differ considerably based on the type of electric vehicle (EV) and its usage, with smaller vehicles typically being more cost-effective at shorter distances [19]. High initial costs, especially for batteries, continue to be a significant barrier to their adoption. However, as technology advances, electric vehicles (EVs) are likely to become more cost-effective, particularly when considering full life-cycle costs, including maintenance and fuel savings [20]. The convenience of charging and accessibility of charging infrastructure are key factors in the adoption of electric vehicles (EVs). Home charging, being both cost-effective and convenient, significantly enhances EV ownership, particularly for those who can charge their vehicles at home [21]. The cost-effectiveness of electric vehicles (EVs) increases with usage patterns that take full advantage of their lower operational costs, such as in cases of high annual mileage. However, if the charging access is limited, convenience decreases, potentially raising overall costs because of the need for more frequent charging or reliance on less accessible charging options [22]. The interplay between convenience, accessibility, and the cost of ownership and operation is crucial for shaping the adoption and usage patterns of electric vehicle (EVs) systems. For EVs, convenience is closely tied to the availability of charging infrastructure, ease of access to maintenance services, and seamless integration of technology such as mobile applications for locating charging stations. In this context, accessibility refers to the geographic distribution of these services, which greatly impacts the overall user experience. When charging stations are plentiful and strategically located, the convenience of using EVs is enhanced, potentially mitigating the higher initial costs compared with traditional internal combustion engine vehicles. Research suggests that the lower operational costs associated with EVs, such as fuel savings, can increase their attractiveness, especially when paired with accessible and convenient infrastructure [23].

The adoption of electric vehicles (EVs) and mass rapid transit systems in urban settings has profound effects on urban infrastructure and transportation networks. Transitioning to EVs plays a crucial role in lowering emissions, enhancing air quality, and fostering sustainable urban development [24]. In densely populated cities, electric transport systems not only provide environmental benefits but also increase the economic viability of infrastructure investments [25]. Additionally, the widespread use of EVs is pivotal in curbing greenhouse gas emissions, particularly in metropolitan areas, where transportation significantly contributes to overall emissions [26]. Integrating mass rapid transit trains with the expansion of EVs can further improve urban mobility by offering efficient and eco-friendly transportation alternatives, thereby alleviating traffic congestion and enhancing accessibility for urban residents [27]. The costs of owning and operating electric vehicles are strongly related to convenience and accessibility. As charging convenience and vehicle range increase, the cost-effectiveness of EVs also increases, making them a more practical choice for a broader spectrum of drivers.

Hypothesis H2. 

Convenience & accessibility have a positive and significant effect on cost of ownership & usage.

2.4.3. Convenience & Accessibility and Travel Patterns & Usage Behavior

Electric vehicle (EV) drivers often exhibit cautious charging behavior, typically charging during the daytime and opting for shorter intervals between charging sessions. This pattern is influenced by the availability of the charging infrastructure and the perceived need to maintain the battery at a high state of charge [28]. The availability of fast chargers can greatly impact travel patterns by enhancing the practicality of electric vehicles (EVs), especially for long-distance trips. Simulations indicate that strategically placing fast chargers can minimize the inconvenience of detours for charging and improve overall travel efficiency [29]. The availability of public charging infrastructure plays a crucial role in shaping electric vehicle (EV) usage patterns. For example, households with convenient access to charging stations are more inclined to use their EVs for longer journeys, which in turn affects their daily travel behavior [30]. In regions with robust charging networks, electric vehicle (EV) drivers are more willing to travel longer distances, and their usage patterns begin to mirror those of conventional vehicles. However, inadequate charging infrastructure tends to confine EV usage to shorter, more predictable routes [31]. Electric vehicle (EV) drivers frequently modify their driving habits over time to accommodate the unique demands of EV operation such as range constraints and charging access. This adaptation can result in more efficient driving behaviors, such as smoother acceleration and braking, which can help extend the range of the vehicles [32]. In multi-car households, electric vehicles (EVs) are often used in a more deliberate manner, reserved for specific trips where the EV’s range is adequate. This emphasizes the critical role of convenience in shaping travel behavior [33]. The convenience and accessibility of public EV charging stations are critical factors influencing the travel patterns of electric vehicle users. Research indicates that the accessibility of these facilities plays a significant role in shaping individuals’ intentions to adopt EVs, with objective, perceived, and prospective accessibility being crucial determinants [10]. Moreover, the interaction between rest and charging behavior during intercity travel can mitigate the constraints imposed by EV battery capacities, underscoring the importance of a comprehensive approach to accessibility [14]. Additionally, the spatial distribution of EV charging infrastructure across various land use areas affects travel behavior, highlighting the necessity of widespread fast-charging options to ensure equitable access across all socioeconomic groups, not just the more affluent [34]. EV adoption and travel patterns can be positively influenced by improving convenience and accessibility through strategic infrastructure planning and addressing disparities in charging access, thereby fostering more sustainable transportation practices. Convenience and access to charging infrastructure significantly impact the travel patterns and usage behavior of electric vehicle (EV) drivers. Comprehensive charging networks can promote longer trips and greater flexibility in usage, whereas limited access tends to confine EV use to shorter, more predictable routes. As the charging infrastructure expands, the EV usage mirrors that of conventional vehicles.

Hypothesis H3. 

Convenience & accessibility have a positive and significant effect on travel patterns & usage behavior.

2.4.4. Technology Perception & Performance and Cost of Ownership & Usage

Consumers frequently view electric vehicles (EVs) as having a higher total cost of ownership (TCO) primarily because of their elevated initial purchase price, even though they may offer long-term savings through reduced operating costs [35]. Precise TCO estimations are essential, as real-world driving conditions can reveal that, in specific scenarios, EVs may have a lower TCO than diesel vehicles [36]. The cost-effectiveness of electric vehicles (EVs) improves with increasing driving distances and is generally more advantageous for smaller vehicles than for larger ones [18]. Government subsidies and incentives are crucial for enhancing the cost-competitiveness of electric vehicles (EVs) relative to internal combustion engine vehicles (ICEVs) and hybrid electric vehicles (HEVs) [37]. The high cost of batteries and concerns regarding their longevity are significant obstacles to the widespread adoption of electric vehicles (EVs). Regular use of fast charging can degrade the battery life, which in turn affects the overall total cost of ownership (TCO) [38]. Public perceptions of electric vehicle (EV) performance, particularly regarding safety, reliability, and driving range, strongly influence consumer acceptance and adoption rates. Consumers who place high value on vehicle performance are generally less inclined to embrace EVs. The widespread adoption of electric vehicles (EVs) and rapid mass trains has substantial implications for urban transportation infrastructure and planning strategies. The uptake of EVs is influenced by various factors, including access to green spaces, parking availability, and loan accessibility, with urban conditions and travel patterns playing pivotal roles [39]. Efficient public EV charging infrastructure planning requires meticulous spatial analysis because poorly located charging stations may discourage potential buyers. An increase in EV usage can transform urban development and alter residents’ travel behaviors, thereby affecting energy infrastructure coordination and fostering sustainable urbanization. Proactive policy initiatives are essential for integrating the spatial distribution of charging stations with sustainable mobility efforts, particularly in developing countries. Furthermore, EV adoption is driven by environmental concerns, economic savings, and income levels, underscoring the need for customized financial incentives and infrastructure development to accelerate the transition to EVs and mitigate the carbon footprint of urban mobility [40]. The perception and performance of electric vehicle (EV) technology are intimately related to the costs of ownership and operation. Although EVs have the potential to offer lower long-term costs, their high upfront purchase prices and battery-related issues remain major challenges. Government incentives and progress in battery technology are essential for reducing the total cost of ownership (TCO) and fostering broader adoption. In addition, regional variations underscore the importance of tailored strategies to effectively promote EV usage.

Hypothesis H4. 

Technology perception & performance have a positive and significant effect on cost of ownership & usage.

2.4.5. Technology Perception & Performance and Travel Patterns & Usage Behavior

Conceptual framework. This study integrates the Technology Acceptance Model and Theory to explain electric-vehicle (EV) purchase intention. The framework posits that charging accessibility and perceived cost shape perceived usefulness and ease of use; environmental concern and subjective norms inform attitudes toward EVs; and these belief-based constructs jointly predict purchase intention, controlling for relevant demographics. Directed arrows indicate hypothesized causal paths.

The drivers’ perceptions of the environmental features of hybrid and electric vehicles play a crucial role in shaping their driving behavior. For example, those who actively utilize features such as regenerative braking or electric propulsion often achieve better fuel efficiency and adjust their driving habits to optimize energy consumption. This suggests a close connection between how drivers perceive and engage with EV technology and their overall driving patterns [41]. Driving behavior, especially in terms of acceleration and braking, significantly affects the energy consumption and range of electric vehicles. Aggressive driving tends to increase energy usage, thereby diminishing the effective range of vehicles. Consequently, this behavior can shape the overall EV usage patterns, as drivers may modify their habits to optimize battery life and efficiency [42]. The adoption of electric vehicles is associated with shifts in travel behaviors and patterns. Research indicates that EV drivers often modify their routes and charging routines based on the availability of the charging infrastructure, which subsequently influences their overall travel patterns. The introduction of fast-charging stations can alleviate the inconvenience of longer trips, thereby affecting the travel choices and habits of EV users [28]. The relationship between technology perception, performance, travel patterns, and usage behavior among electric vehicle (EV) users is intricate and shaped by factors such as user experience, driving habits, and environmental awareness. Users’ perceptions of electric vehicles, particularly regarding range satisfaction and charging convenience, play a crucial role in shaping their travel patterns and usage behaviors. Research indicates that as users become more familiar with EVs, their satisfaction with the range and charging ease tends to grow, leading to greater acceptance and more frequent use of EVs [43]. Initial experience with EVs can alter driving habits. Research shows that after an initial adjustment period, drivers often adopt a calmer driving style with EVs compared with traditional internal combustion engine vehicles. This shift may be attributed to the unique performance characteristics of EVs, such as the instant torque and regenerative braking [32]. Battery Electric Vehicle (BEV) users are generally male, well educated, and affluent, often with multiple vehicles in their households. They encounter fewer functional barriers and hold more favorable attitudes toward BEVs than conventional vehicle (CV) users. The study suggests that distinct factors influence BEV adoption for each group, with symbolic attitudes crucial for CV users, whereas driving range concerns are more significant for BEV users. In households with multiple cars, the actual use of BEVs is influenced by the type of other vehicles, perceived barriers, and successful adaptation to longer trips [44]. The drivers’ perceptions of EV technology and its performance have a direct impact on their driving behavior and travel patterns. As drivers gain familiarity with the capabilities and limitations of EVs, they tend to adjust their travel behavior, resulting in a more efficient use of the vehicle’s features and potentially altering their overall travel habits. In sum, the framework links infrastructure- and belief-based antecedents to EV purchase intention through perceived usefulness, ease of use, risk, attitude, and norms. The following hypotheses are derived from these directional relationships and empirically evaluated in the next section.

Hypothesis H5. 

Technology perception & performance have a positive and significant effect on travel patterns & usage behavior.

2.4.6. Cost of Ownership & Usage and Travel Patterns & Usage Behavior

The interplay between ownership costs, usage behavior, and travel patterns among electric vehicle (EV) drivers is shaped by multiple factors such as charging infrastructure, trip distances, and vehicle attributes. Studies have demonstrated that, particularly when factoring in incentives and grants, the cost of owning an EV can be substantially lower than that of internal combustion engine vehicles (ICEVs) over a medium-term ownership period [45]. Travel patterns significantly influence cost dynamics, as battery electric vehicles (BEVs) are predominantly used for shorter trips, with the majority of journeys being under 15 km, making them well-suited for urban settings [46]. While cost-efficient EVs are becoming more accessible, user behavior such as daily trip lengths and charging practices remain essential for optimizing usage patterns and minimizing overall ownership costs [47]. The frequency and distance of electric vehicle usage are heavily influenced by the cost of electricity and accessibility of the charging infrastructure. In regions where electricity is cheaper or charging stations are more readily available, drivers typically use EVs more frequently. This indicates that lower operational costs promote increased usage, leading to different travel patterns compared to those seen in traditional internal combustion engine vehicles [48]. Electric vehicle usage patterns can differ greatly depending on the costs of ownership and operation. In regions where electricity is expensive or charging infrastructure is scarce, EV drivers may adopt more conservative behaviors such as taking shorter trips and charging less frequently. Conversely, in areas with lower costs and better infrastructure, drivers are likely to take longer trips and charge more often [28]. The relationship between ownership costs and usage behavior in electric vehicles (EVs) is complex and shaped by various factors, such as vehicle range, charging infrastructure, and individual driving habits. The economics of Battery Electric Vehicles (BEVs) are closely tied to the driving patterns, vehicle range, and charging strategies. Differences in daily driving distances, charging behaviors, and battery degradation play a significant role in determining the total cost of ownership and overall economic viability of EVs [49]. Research indicates that enabling users to choose EVs tailored to their specific travel patterns can decrease their reliance on fast chargers and the associated energy consumption, thereby optimizing the balance between range and cost. This underscores the significance of personalized vehicle ranges in minimizing the overall ownership costs [50]. Socioeconomic factors, such as income and education, significantly influence ownership patterns and travel behaviors, affecting EV adoption and usage. For example, wealthier households are more likely to adopt EVs and their travel patterns typically involve greater daily distances, which in turn affects their charging and usage behaviors [51]. The incorporation of EVs into daily usage patterns impacts energy demand, especially in urban areas, where charging infrastructure is a significant concern. In some regions, the ability to charge vehicles at home overnight and discharge them during the day can help optimize electricity demand and lower costs [52]. The costs of ownership and operation have a direct impact on the travel patterns and behavior of electric vehicle drivers. Reduced costs, particularly in terms of energy and maintenance, promote more frequent and longer use of EVs, resulting in changes to travel behavior and patterns. The intricate relationship between travel behavior and vehicle costs underscores the need for strategic policies and infrastructure investments to promote widespread EV adoption and foster sustainable transportation systems.

Hypothesis H6. 

Cost of ownership & usage have a positive and significant effect on travel patterns & usage behavior.

Convenience and accessibility have a positive and significant effect on travel patterns and usage behavior using technological perception and performance as mediator variables.

The connection between convenience and accessibility and travel patterns and usage behavior among electric vehicle (EV) drivers is heavily mediated by technological perception and performance. Research indicates that perceived ease of use and performance of EVs play a crucial role in shaping user behavior and travel patterns. For example, a study on tourist perceptions in the Delhi National Capital Region revealed that improving technological features such as ease of use and performance boosts user satisfaction and adoption intentions, thereby influencing travel patterns by making EVs more accessible and convenient for users [53]. Furthermore, research in Saudi Arabia demonstrates that technological innovativeness, which shapes perceptions of EVs’ utility and sustainability, also mediates the link between accessibility and usage behavior. This underscores the critical role of technological advancement in promoting wider EV adoption [54]. Nonetheless, challenges persist, especially in areas with lower EV enthusiasm, where consumer attitudes and perceptions play a significant role in influencing adoption intentions [55]. Accessibility analysis shows that in regions with advanced charging infrastructure, such as Beijing-Tianjin-Hebei, the performance of battery electric vehicles (BEVs) can rival that of traditional fuel vehicles, reducing concerns about travel convenience [14]. Collectively, the analyses outlined above provide a consistent basis for testing the proposed relationships using appropriate controls and robustness checks. We now report the empirical results and assess each hypothesis against the specified criteria.

Hypothesis H7. 

Convenience & accessibility have a positive and significant effect on travel patterns & usage behavior by technological perception and performance is a mediator variable.

Convenience and accessibility have positive and significant effects on travel patterns and usage behavior because the cost of ownership and usage is a mediator variable.

The interplay between convenience, accessibility, and travel patterns is closely connected to the usage behavior of electric vehicle (EV) drivers, with the cost of ownership and usage acting as a pivotal mediator. Variables related to convenience and accessibility, such as the presence of charging stations and ease of parking, significantly influenced travel patterns by determining the feasibility of using EVs for different types of trips. These factors also affect usage behavior, including the frequency of EV use, distance traveled, and route selection. However, the relationship between convenience, accessibility, and travel patterns is moderated by the cost of ownership and usage, which includes considerations such as initial purchase price, maintenance expenses, and energy costs. For example, a well-established charging infrastructure might promote more frequent EV use, but high ownership and maintenance costs could diminish these advantages, resulting in more cautious usage behavior [56]. Turning to the broader context, we elaborate how these results inform EV adoption strategies. The transportation sector is widely acknowledged as a significant contributor to emissions, making electric vehicles (EVs) a promising solution to address this challenge. Despite the promotion of EVs as an optimal alternative, encouraging their widespread adoption remains a significant challenge. This study contributes to this area of inquiry by investigating the influence of self-identity on EV purchase intentions and behavior among millennials in Vietnam, a rapidly developing market. Self-identity has both direct and indirect effects on EV purchase behavior, with purchase orientation serving as a mediator and UTAUT variables such as consumer expectations and motivations acting as moderators [57].

Hypothesis H8. 

Convenience & accessibility have a positive and significant effect on travel patterns & usage behavior by cost of ownership and usage is a mediator variable.

After analyzing the literature on factors influencing electric vehicle user behavior, a conceptual framework for the research was developed, outlining the interconnections between each variable. The model is grounded in the theoretical frameworks of Behavioral Economics, Consumer Decision-Making, and the Technology Acceptance Model (TAM). It utilizes structural equation modeling (SEM) to examine and interpret the relationships between these concepts.

Convenience and Accessibility (CA): Convenience and Accessibility (CA), an exogenous latent variable, impacts both Technological Perception and Performance (TP) as well as the Cost of Ownership and Usage (CU). This variable includes factors such as ease of use, travel time, service frequency, and accessibility, which collectively influence the travel patterns and usage behaviors of EV users.

Technological Advancements (TP), a latent construct capturing users’ evaluations of technological usefulness, ease of use, and performance reliability, is an endogenous variable that encompasses perceptions of reliability, performance, innovation, and technological progress. It is influenced by convenience and accessibility, and in turn, impacts the cost of ownership and usage, as well as travel patterns and usage behavior.

Cost of Ownership and Usage (CU), an endogenous variable, acts as a mediator between convenience and accessibility and travel patterns and usage behavior, as well as between technological perception and performance. This variable encompasses factors such as initial purchase price, maintenance expenses, ticket prices for rapid trains, and cost savings.

Travel Patterns and Usage Behavior (TU) are the main endogenous variables in the model and are directly affected by convenience and accessibility, technological perception, and cost of ownership and usage. This serves as the model’s outcome, reflecting consumers’ travel patterns and usage behavior regarding EVs.

3. Research Methodology

3.1. Research Design

This exploratory research begins with an in-depth literature review that examines the factors shaping consumer behavior in the adoption of electric vehicles, with a focus on identifying the underlying variables. This study seeks to uncover the relationships between various elements that influence consumer decisions regarding electric vehicle usage. It explores the interplay between convenience and accessibility, technological perception and performance, and the cost of ownership and usage, and assesses their combined impact on travel patterns and usage behavior. Additionally, this research delves into how these factors, both individually and collectively, affect the travel behaviors and usage patterns of electric vehicle drivers.

3.2. Data Collection Process

This research employed a quantitative methodology, with questionnaires serving as the primary tool for data collection. The questionnaire was structured into two key sections. The first section collected demographic data from electric vehicle drivers, while the second section investigated the relationships between convenience and accessibility, technological perception and performance, and the cost of ownership and usage, and how these factors influence travel patterns and usage behavior among electric vehicle drivers. To assess driver behavior, a Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree) was used, focusing on the aspects of convenience and accessibility, technological perception and performance, cost of ownership and usage, and travel patterns and usage behavior. The gathered data were statistically analyzed to interpret the research findings and survey outcomes.

Beyond the Likert-scale questions, the survey incorporated multiple-choice and open-ended questions to gather a wider range of responses and deeper insights into travel patterns and usage behaviors. The multiple-choice questions primarily addressed the respondents’ demographic details, including age, education level, income, and geographic location. The open-ended questions provided participants with the opportunity to elaborate on aspects related to convenience and accessibility, technological perception and performance, and the cost of ownership and usage, as they pertained to their travel patterns and usage behavior.

The survey employed a mixed-methods approach, combining online distribution with face-to-face interviews to achieve thorough data collection. The online survey was disseminated via social media platforms and email lists aimed at individuals interested in the travel patterns and usage behavior of electric vehicles. Face-to-face interviews were conducted in diverse locations encompassing both urban and rural areas to ensure a representative sample across various demographic groups.

The sample was gathered using a stratified sampling technique to align closely with the general population of Thailand. Stratification was based on key demographic factors, such as age, education level, and income. This method ensured that the sample more accurately represented the Thai population in these crucial areas, allowing the findings to be generalized to a broader population with greater confidence. Additionally, post-survey weighting was applied to correct any discrepancies between the sample and general population.

Sampling frame and quotas. We targeted adults eligible to purchase or use EVs and enforced soft quotas by residence (urban/rural) and EV ownership (owner/non-owner) to reflect the market composition. The achieved sample was n = 398 (urban = 72%, rural = 28%; owners = 86%, non-owners = 14%). Post-stratification weighting. To improve alignment with population margins, we computed weights using iterative proportional fitting (raking) on the two dimensions (residence, ownership). Analyses report weighted estimates, with unweighted results shown in robustness checks; patterns are materially unchanged.

Sensitivity checks. We (a) compare key coefficients across weighted and unweighted models, (b) run a multi-group SEM (urban vs. rural; owners vs. non-owners), and (c) report measurement invariance tests across these groups (configural/metric/scalar). The results were consistent across the groups.

We employed a stratified design, with strata defined by residence (urban vs. rural) and EV ownership status (owner vs. non-owner). The sampling frame comprised adults eligible to purchase or use EVs within the next 3–5 years. Target allocations reflect market composition and analytic needs (oversampling owners to secure power for subgroup tests). Within each stratum, h, respondents were selected using random digit/online panel draws with eligibility screening and no replacement. Final completes by stratum were: urban-owner nUO = [ ], urban-non-owner nUN = [ ], rural-owner nRO = [ ], rural-non-owner nRN = [ ] (total = 398).

3.3. Data Analysis

The data analysis process commenced with data cleaning, where multivariate outliers were identified using the Mahalanobis Distance method in SPSS Version 26. Outliers with a p-value less than 0.001 were excluded from further analysis. Of the initial 400 samples, two outliers were detected, resulting in 398 samples being used for subsequent analysis. Construct validity was evaluated by examining factor-loading values, with a threshold of 0.4 or higher. The reliability of the questionnaire was assessed using Cronbach’s alpha, with values above 0.7 indicated acceptable reliability. After these preliminary steps, the data were analyzed using descriptive statistics, followed by structural equation modeling (SEM) using AMOS statistical software.

The structural equations that outline the mathematical relationships between the variables in the proposed structural equation model (SEM) for the travel patterns and usage behavior of EVs, as illustrated in Figure 2, are presented below.

Figure 2 illustrates the mathematical structures corresponding to these structural equations. The structural equation model (SEM) diagram depicts the relationships between convenience and accessibility (ξ_CA), technological perception and performance (η_TP), cost of ownership and usage (η_CU), and travel patterns and usage behavior (η_TU). In this model, each construct is represented by latent variables (circles) and observed variables (rectangles). Error terms (ε and ζ) are also included to account for measurement errors in both the observed and latent variables, with the structural equations outlined in Equation (1).

Regression-style “where” clause (fits OLS/GLS with latent scores)

where:

YYY = dependent outcome [e.g., purchase intention, specify, for example, purchase intention measured on a Likert1–5 scale;

CA = Convenience and Accessibility (latent score; higher = greater convenience/charging access)

TP = Technology Perception and Performance (latent score; higher = more favorable tech beliefs);

CU = Cost of Ownership and Usage (latent score; higher = lower perceived cost/greater affordability, reverse-coded if applicable);

X = vector of controls [e.g., age, income, priorEVexperience];

β0\beta_0β0 = intercept; β1, β2, β3 = slope coefficients (standardized unless noted);

[ε] = disturbance term, E[ε] = 0E [ε] = 0, Var(ε) = σ2

Expected signs: β1 > 0, β2 > 0, β3 > 0 [adjust if different]

Structural Equation for Path Coefficient

$$\eta =\beta \eta +\Gamma \xi +\zeta$$

(1)

$$\left[\begin{array}{c}{\mathsf{\eta }}_{TP}\\ {\mathsf{\eta }}_{CU}\\ {\mathsf{\eta }}_{TU}\end{array}\right]=\left[\begin{array}{ccc}0& 0& 0\\ {\mathsf{\beta }}_1& 0& 0\\ {\mathsf{\beta }}_2& {\mathsf{\beta }}_3& 0\end{array}\right]\left[\begin{array}{c}{\mathsf{\eta }}_{TP}\\ {\mathsf{\eta }}_{CU}\\ {\mathsf{\eta }}_{TU}\end{array}\right]+\left[\begin{array}{c}{\mathsf{\gamma }}_1\\ {\mathsf{\gamma }}_2\\ {\mathsf{\gamma }}_3\end{array}\right]\left[\begin{array}{c}{\mathsf{\xi }}_{CA}\end{array}\right]+\left[\begin{array}{c}{\mathsf{\zeta }}_1\\ {\mathsf{\zeta }}_2\\ {\mathsf{\zeta }}_3\end{array}\right]$$

(2)

Structural Equation for Endogenous Variables

$$y={\Lambda }_y\eta +\epsilon$$

(3)

$$\left[\begin{array}{c}{\mathrm{y}}_{TP1}\\ {\mathrm{y}}_{TP2}\\ {\mathrm{y}}_{TP3}\\ {\mathrm{y}}_{TP4}\\ {\mathrm{y}}_{CU1}\\ {\mathrm{y}}_{CU2}\\ {\mathrm{y}}_{CU3}\\ {\mathrm{y}}_{CU4}\\ {\mathrm{y}}_{TU1}\\ {\mathrm{y}}_{TU2}\\ {\mathrm{y}}_{TU3}\\ {\mathrm{y}}_{TU4}\end{array}\right]=\left[\begin{array}{ccc}{\lambda }_{TP1}^y& 0& 0\\ {\lambda }_{TP2}^y& 0& 0\\ {\lambda }_{TP3}^y& 0& 0\\ {\lambda }_{TP4}^y& 0& 0\\ 0& {\lambda }_{CU1}^y& 0\\ 0& {\lambda }_{CU2}^y& 0\\ 0& {\lambda }_{CU3}^y& 0\\ 0& {\lambda }_{CU4}^y& 0\\ 0& 0& {\lambda }_{TU1}^y\\ 0& 0& {\lambda }_{TU2}^y\\ 0& 0& {\lambda }_{TU3}^y\\ 0& 0& {\lambda }_{TU4}^y\end{array}\right]\left[\begin{array}{c}{\mathsf{\eta }}_{TP}\\ {\mathsf{\eta }}_{CU}\\ {\mathsf{\eta }}_{TU}\end{array}\right]+\left[\begin{array}{c}{\mathit{\epsilon }}_{TP1}\\ {\mathit{\epsilon }}_{TP2}\\ {\mathit{\epsilon }}_{TP3}\\ {\mathit{\epsilon }}_{TP4}\\ {\mathit{\epsilon }}_{CU1}\\ {\mathit{\epsilon }}_{CU2}\\ {\mathit{\epsilon }}_{CU3}\\ {\mathit{\epsilon }}_{CU4}\\ {\mathit{\epsilon }}_{TU1}\\ {\mathit{\epsilon }}_{TU2}\\ {\mathit{\epsilon }}_{TU3}\\ {\mathit{\epsilon }}_{TU4}\end{array}\right]$$

(4)

Structural Equation for Exogenous Variables

$$x={\Lambda }_x\xi +\delta$$

(5)

$$\left[\begin{array}{c}{\mathrm{x}}_{CA1}\\ {\mathrm{x}}_{CA2}\\ {\mathrm{x}}_{CA3}\\ {\mathrm{x}}_{CA4}\end{array}\right]=\left[\begin{array}{c}{\lambda }_{CA1}^x\\ {\lambda }_{CA2}^x\\ {\lambda }_{CA3}^x\\ {\lambda }_{CA4}^x\end{array}\right]\left[\begin{array}{c}{\mathsf{\xi }}_{CA}\end{array}\right]+\left[\begin{array}{c}{\delta }_{CA1}\\ {\delta }_{CA2}\\ {\delta }_{CA3}\\ {\delta }_{CA4}\end{array}\right]$$

(6)

4. Results

4.1. Demographic Information of Electric Vehicles Users

Table 1. Demographic results.

Items	Details	Frequency	Percentage
Gender	Male	109	27.4
	Female	289	72.6
Age	20–29 years	81	20.4
	30–39 years	125	31.4
	40–49 years	114	28.6
	More than 49 years	78	19.6
Income USD (monthly)	215–515	30	7.6
	516–828	142	35.8
	829–1142	134	33.8
	More than 1142	91	22.8
Education level	Lower than bachelor’s degree	57	14.3
	Bachelor’s degree	212	53.2
	Master’s degree	114	28.7
	Higher than master’s degree	15	3.8

presents the demographic characteristics of the respondents. Most respondents were female (n = 289; 72.6%), while 27.4% were male (n = 109). Gender parity was not a quota in our stratified design; the realized composition reflects the panel availability for EV-interest screening. Regarding age distribution, 31.4% of the respondents fell within the 30–39 years age bracket (n = 125), followed by 28.6% aged 40–49 years (n = 114), 20.4% aged 20–29 years (n = 81), and 19.6% aged over 49 years (n = 78). In terms of monthly income, 35.8.% reported earning between USD 516 and 830 (n = 142), 33.8% earned between USD 831 and 1143 (n = 134), 22.8% earned more than USD 1144 (n = 91), and 19.6% earned between USD 215 and 515 (n = 30). Regarding educational attainment, 53.2% had a bachelor’s degree (n = 212), 28.7% had a master’s degree (n = 114), 14.3% had education levels below a bachelor’s degree (n = 57), and 3.8% had education levels higher than a master’s degree (n = 15).

Consumer segmentation and heterogeneity. To examine whether key relationships differ across demographic segments, we estimated the multi-group SEM over predefined groups for age (≤35, 36–50, >50), household income (tertiles), and education (≤secondary/tertiary+). Before comparing paths, we assessed measurement invariance for all latent constructs across groups (configural → metric → scalar). Invariance was judged using ΔCFI ≤ 0.01 (with ΔRMSEA/ΔSRMR as auxiliaries). Upon establishing at least metric invariance, we test structural path differences by (i) constraining focal paths to equality, and (ii) conducting Wald/Δχ² comparisons against an unconstrained model. Reported effects are standardized; SEs use MLR (robust) with survey weights, where applicable. As a complementary approach, we estimate an exploratory latent class SEM (two to four classes) using information criteria (AIC/BIC) and Lo–Mendell–Rubin tests to detect unobserved heterogeneity. Class-specific paths are reported with posterior means as descriptive profiles (age/income/education). Finally, we probe moderation with interaction terms (e.g., Convenience & Accessibility × high income) using product-indicator methods; the results mirror multi-group comparisons.

4.2. Reliability Testing

In the statistical analysis, Cronbach’s Alpha was utilized to evaluate the reliability of the survey data coefficients. As shown in

Table 2. Convergent validity, discriminant validity, and reliability results.

Construct	Variables	Factor Loading	CR	AVE	Cronbach’s Alpha
Convenience and accessibility	Convenience of use	0.697	0.849	0.586	0.856
	Travel time	0.712
	Frequency of service	0.777
	Ease of access	0.865
Technological perception and performance	Perceptions of reliability	0.735	0.816	0.526	0.816
	Performance	0.751
	Technological advancements	0.697
	Innovation	0.717
Cost of ownership and usage	Initial purchase price	0.754	0.819	0.531	0.829
	Maintenance costs	0.764
	Ticket prices for rapid train	0.677
	Cost savings	0.716
Travel patterns and usage behavior	Frequency of use	0.778	0.845	0.577	0.838
	Trip purposes (commuting vs. leisure)	0.699
	Distance traveled	0.837
	Multimodal travel behavior	0.717

, Cronbach’s alpha values ranged from 0.816 to 0.856, exceeding the recommended minimum threshold of 0.7 for all latent variables. These results confirmed the high reliability and internal consistency of the survey measurements.

Convergent validity analysis, as shown in Table 2, demonstrated that both observed and latent variables met the criteria for convergent validity. This was indicated by (1) all observed variables with statistically significant factor loadings ranging from 0.697 to 0.865, well above the 0.5 threshold, and (2) convergent validity surpassing the cutoff value of 0.6.

Table 2 presents the average variance extracted (AVE). The results showed that the AVE values for all latent variables ranged from 0.531 to 0.586, meeting the discriminant validity criterion of exceeding 0.5.

Structural Equation for Path Coefficient

$$\eta =\beta \eta +\Gamma \xi +\zeta$$

(7)

$$\left[\begin{array}{c}{\mathsf{\eta }}_{TP}\\ {\mathsf{\eta }}_{CU}\\ {\mathsf{\eta }}_{TU}\end{array}\right]=\left[\begin{array}{ccc}0& 0& 0\\ 0.213& 0& 0\\ 0.325& 0.203& 0\end{array}\right]\left[\begin{array}{c}{\mathsf{\eta }}_{TP}\\ {\mathsf{\eta }}_{CU}\\ {\mathsf{\eta }}_{TU}\end{array}\right]+\left[\begin{array}{c}0.342\\ 0.343\\ 0.179\end{array}\right]\left[\begin{array}{c}{\mathsf{\xi }}_{CA}\end{array}\right]+\left[\begin{array}{c}0.883\\ 0.787\\ 0.707\end{array}\right]$$

(8)

$${\mathsf{\eta }}_{TP}=0.342{\mathsf{\xi }}_{CA}+0.883$$

(9)

$${\mathsf{\eta }}_{CU}=0.213{\mathsf{\eta }}_{TP}+0.343{\mathsf{\xi }}_{CA}+0.787$$

(10)

$${\mathsf{\eta }}_{TU}=0.325{\mathsf{\eta }}_{TP}+0.203{\mathsf{\eta }}_{CU}+0.179{\mathsf{\xi }}_{CA}+0.707$$

(11)

Structural Equation for Endogenous Variables

$$y={\Lambda }_y\eta +\epsilon$$

(12)

$$\left[\begin{array}{c}{\mathrm{y}}_{TP1}\\ {\mathrm{y}}_{TP2}\\ {\mathrm{y}}_{TP3}\\ {\mathrm{y}}_{TP4}\\ {\mathrm{y}}_{CU1}\\ {\mathrm{y}}_{CU2}\\ {\mathrm{y}}_{CU3}\\ {\mathrm{y}}_{CU4}\\ {\mathrm{y}}_{TU1}\\ {\mathrm{y}}_{TU2}\\ {\mathrm{y}}_{TU3}\\ {\mathrm{y}}_{TU4}\end{array}\right]=\left[\begin{array}{ccc}0.735& 0& 0\\ 0.751& 0& 0\\ 0.697& 0& 0\\ 0.717& 0& 0\\ 0& 0.754& 0\\ 0& 0.764& 0\\ 0& 0.677& 0\\ 0& 0.716& 0\\ 0& 0& 0.778\\ 0& 0& 0.699\\ 0& 0& 0.837\\ 0& 0& 0.717\end{array}\right]\left[\begin{array}{c}{\mathsf{\eta }}_{TP}\\ {\mathsf{\eta }}_{CU}\\ {\mathsf{\eta }}_{TU}\end{array}\right]+\left[\begin{array}{c}0.460\\ 0.435\\ 0.515\\ 0.486\\ 0.432\\ 0.417\\ 0.542\\ 0.488\\ 0.395\\ 0.511\\ 0.300\\ 0.487\end{array}\right]$$

(13)

Structural Equation for Exogenous Variables

$$x={\Lambda }_x\xi +\delta$$

(14)

$$\left[\begin{array}{c}{\mathrm{x}}_{CA1}\\ {\mathrm{x}}_{CA2}\\ {\mathrm{x}}_{CA3}\\ {\mathrm{x}}_{CA4}\end{array}\right]=\left[\begin{array}{c}0.697\\ 0.712\\ 0.777\\ 0.865\end{array}\right]\left[\begin{array}{c}{\mathsf{\xi }}_{CA}\end{array}\right]+\left[\begin{array}{c}0.514\\ 0.493\\ 0.396\\ 0.252\end{array}\right]$$

(15)

4.3. Structural Equation Analysis (SEM Analysis)

Figure 3 illustrates the path analysis diagram of the proposed research model. Statistical analysis of this model revealed a χ² value of 97.61 (df = 77, p = 0.05), indicating statistical significance. The relative chi-square value of 1.268 falls within the acceptable range of 0 to 2, suggesting a good model fit, as per [58]. Various model fit indices were calculated: RMSEA = 0.026, NFI = 0.966, and GFI = 0.972. An RMSEA value of less than 0.05 indicates an excellent fit for the model. These results confirm the validity of the proposed model for understanding the travel patterns and usage behaviors of electric vehicle users.

The details of the coefficient values in Figure 3 come from the path analysis and structural equation modeling (SEM) results discussed in this study. A summary of the key coefficient values and their significance is presented in this study. Path coefficient values: These values represent the strength and direction of relationships between latent variables (e.g., convenience and accessibility, technology perception and performance, cost of ownership and usage, and travel patterns and usage behavior).

Convenience and accessibility (CA) → technology perception and performance (TP): path coefficient = 0.342, p < 0.001 (supported at the 0.001 level).

Convenience and accessibility (CA) → cost of ownership and usage (CU): path coefficient = 0.343, p < 0.001 (supported at the 0.001 level).

Convenience and accessibility (CA) → travel patterns and usage behavior (TU): path coefficient = 0.179, p < 0.003 (supported at the 0.01 level).

Technology perception and performance (TP)→ cost of ownership and usage (CU): path coefficient = 0.213, p < 0.001 (supported at the 0.001 level).

Technology perception and performance (TP)→ travel patterns and usage behavior (TU): path coefficient = 0.325, p < 0.001 (supported at the 0.001 level).

Cost of ownership and usage (CU)→ travel patterns and usage behavior (TU): path coefficient = 0.203, p-value = 0.002 (supported at the 0.01 level).

Mediation analysis: This study also performed a mediation analysis to explore the indirect effects through intermediate variables.

Convenience and accessibility (CA) → technology perception and performance (TP) → travel patterns and usage behavior (TU): indirect effect = 0.094, p-value = 0.003 (partial mediation). Convenience and accessibility (CA) → Cost of ownership and usage (CU)→ travel patterns and usage behavior (TU): indirect effect = 0.059, p-value = 0.005 (partial mediation).

Overall model fit: The SEM analysis provided fit indices that suggested a good model fit: chi-square = 97.613, df = 77, p-value = 0.056. RMSEA = 0.026, RMR = 0.013, GFI = 0.972, NFI = 0.966, TLI = 0.988, CFI = 0.993.

4.3.1. Relationships of Causality Among Latent Variables

The results of the path analysis, as presented in

Table 3. The hypothesis testing.

Hypothesis	Path	Path Coefficient	p-Value	Relationship
H1	CA   TP	0.342 ***	<0.001	Supported
H2	CA   CU	0.343 ***	<0.001	Supported
H3	CA   TU	0.179 **	0.003	Supported
H4	TP   CU	0.213 ***	<0.001	Supported
H5	TP   TU	0.325 ***	<0.001	Supported
H6	CU   TU	0.203 **	0.002	Supported

Note: Sig at 0.05, ** Sig at 0.01, *** Sig at 0.001 level.

, indicate that a path coefficient value of less than one signifies a causal influence of the independent variable on the dependent variable. Accordingly, hypothesis testing demonstrated that convenience and accessibility (CA) factors have a positive and statistically significant effect on technology perception and performance (TP) factors (H1) with a path coefficient of β = 0.342 (p < 0.001). Similarly, CA factors were found to have a positive and significant impact on cost of ownership and usage (CU) factors (H2) with β = 0.343 (p < 0.001), as well as on travel patterns and usage behavior (TU) factors (H3) with β = 0.179 (p < 0.01). Moreover, the technology perception and performance (TP) factors positively and significantly influence the cost of ownership and usage (CU) factors (H4) with β = 0.213 (p < 0.001) and travel patterns and usage behavior (TU) factors (H5) with β = 0.325 (p < 0.001). Finally, cost of ownership and usage (CU) factors were also found to have a positive and significant effect on travel patterns and usage behavior (TU) (H6) with β = 0.203 (p < 0.01).

4.3.2. Mediation Analysis

Table 4. Mediation analysis.

Hypothesis	Paths	Direct Effect	Indirect Effect	p-Value	Mediation	Relationship
H7	CA → TU	0.179 **		0.003	Partial	Supported
	CA → TP → TU		0.094 **	0.003		Supported
H8	CA → TU	0.179 **		0.003	Partial	Supported
	TI → CU → TU		0.059 **	0.005		Supported

Note: ** Sig at 0.01 level.

displays the results of the mediation analysis, which is a key part of structural equation modeling. This analysis explored the indirect impact of the two causal variables through a mediating factor. In this study, technological perception, performance, and factors related to ownership and usage costs serve as mediating variables to examine the connection between convenience and accessibility with travel patterns and usage behavior. The analysis starts by exploring the relationship between travel patterns, usage behavior, convenience, and accessibility, with technology perception and performance as the mediating variable (H7). Following this, the study investigated the relationship between convenience, accessibility, travel patterns, and usage behavior, with travel patterns and usage behavior serving as the mediating variable (H8). For the mediation variable (H7), examining the relationship between convenience and accessibility, travel patterns, and usage behavior, the indirect effect had a coefficient of 0.094, which was statistically significant at the 0.01 level. This suggests that convenience and accessibility factors partially mediate the relationships between these variables.

In the analysis of the mediator variable (H8), which focused on the relationship between convenience, accessibility, travel patterns, and usage behavior, the indirect effect showed a statistically significant coefficient of 0.059 at a 0.01 level. This indicates that travel patterns and usage behavior factors partially mediate the relationship between convenience and accessibility and these behavioral patterns.

5. Discussion

The results strongly support the hypothesis that convenience and accessibility play a crucial role in shaping consumers’ technological perceptions and performance of electric vehicles (EVs). Convenience and accessibility have a positive and significant effect on technology perception and performance (H1). This relationship is primarily driven by factors such as availability of charging infrastructure, ease of use, and user satisfaction with the overall EV experience. Structural equation modeling (SEM) analysis revealed a significant positive path coefficient (β = 0.342, p < 0.001), indicating that as convenience and accessibility improve, so does consumers’ perception of the technology associated with EVs. Accessibility, in terms of both physical infrastructure and perceived ease of use, can enhance user satisfaction and facilitate a smoother adoption process. Therefore, improving access to charging stations and simplifying the use of EV technology can significantly influence consumer perceptions and ultimately promote wider adoption. EV drivers experience greater satisfaction and perceive higher efficiency than conventional vehicles do. This is primarily attributed to the ease of use and perceived technological innovations, which simplify driving and make electric vehicles more appealing [59]. Studies have suggested that both actual and perceived access to EV charging stations play a critical role in shaping individuals’ intentions to adopt and use electric vehicles. Research has demonstrated that increasing the availability of charging stations in residential areas enhances perceived convenience and boosts the probability of EV adoption [10].

Convenience and accessibility factors have positive and significant effects on the cost of ownership and usage factors (H2). The study also finds significant evidence supporting the positive impact of convenience and accessibility on the perceived cost of ownership and usage of EVs. With a path coefficient of β = 0.343 (p < 0.001), the results indicate that improved accessibility to charging infrastructure and convenience of operating EVs reduce perceived operational costs. The total cost of ownership (TCO) is often a primary concern for potential EV adopters, with high initial costs acting as barriers. However, as infrastructure improves and home charging becomes more feasible, these concerns diminish, rendering EVs a more attractive option for a broader range of consumers. Accessible charging and reduction in range anxiety also contribute to long-term savings in maintenance and fuel costs, thereby supporting the overall cost-effectiveness of EVs. The convenience and accessibility of electric vehicle (EV) infrastructure, particularly in terms of charging stations, have a significant and positive effect on the cost of ownership and usage for EV drivers. Studies from various regions, including the UK, the US, and Japan, have revealed that electric vehicles are becoming more cost-competitive than conventional vehicles. This is largely because of enhanced infrastructure and government subsidies, which have reduced the total cost of ownership (TCO), making EVs a more appealing choice for both individual and fleet owners [37].

Convenience and accessibility factors have a positive and significant effect on travel patterns and usage behavior factors (H3). The availability of EV charging infrastructure significantly influences travel patterns and usage behavior. The SEM analysis confirmed this finding, with a positive and significant path coefficient (β = 0.179, p < 0.01). Households with easy access to charging stations are more inclined to use EVs for long-distance trips, and their usage patterns mirror those of conventional vehicles. The role of convenience is highlighted in the way drivers adapt their travel and charging habits, based on the accessibility of public and private charging facilities. Limited access tends to confine EV use to predictable, shorter routes, whereas extensive infrastructure encourages longer trips and greater flexibility. Convenience and accessibility are key determinants of travel patterns and usage behavior among electric vehicle (EV) drivers, as evidenced by studies such as those conducted in the Beijing-Tianjin-Hebei region. This research suggests that beyond a certain threshold, increasing the number of charging stations does not significantly enhance accessibility. Instead, strategically positioning stations and improving charging speed through technological advancements have a greater impact than merely increasing the number of stations [14]. In cities such as Shanghai, EV drivers often make frequent short trips between charging sessions, and many choose to charge their vehicles during the day. Driving and charging behaviors vary significantly and are influenced by individual travel requirements and activity patterns [28].

Technology perception and performance factors have a positive and significant effect on the cost of ownership and usage factors (H4). Technology perception and performance also significantly affect how consumers perceive the costs of owning and using EV. SEM analysis indicated a positive relationship, with a path coefficient of β = 0.213 (p < 0.001). Consumers who perceive EVs as technologically advanced and reliable are more likely to view the cost of ownership and usage favorably, particularly as long-term benefits, such as lower maintenance costs and fuel savings, become evident. Technological advancements, including improved battery performance and more reliable charging infrastructure, reduce concerns about cost and encourage broader adoption. Consumer interest in electric vehicles (EVs) is shaped by their perceived environmental advantages, societal expectations, and individual attitudes toward technological innovation [60]. The perception of technology and performance play a crucial role in influencing the total cost of ownership (TCO) and usage for electric vehicle (EV) drivers. Older consumers and individuals with strong pro-environmental tendencies are more inclined to adopt electric vehicles (EVs), whereas younger consumers and those prioritizing performance exhibit lower levels of acceptance [61]. Although economic incentives and long-term savings can increase the appeal of EVs, ongoing concerns about their performance and reliability continue to pose challenges. By advancing technological solutions, improving charging infrastructure, and effectively communicating the financial advantages of TCO, these concerns can be mitigated, ultimately fostering greater EV adoption.

Technology perception and performance factors have a positive and significant effect on travel patterns and usage behavior factors (H5). The performance and technological features of EVs play a significant role in shaping travel behavior. With a positive path coefficient (β = 0.325, p < 0.001), the study demonstrated that improvements in EV technology, such as increased range and charging speed, lead to more favorable travel patterns and greater usage of EVs. Drivers are more likely to adjust their behavior, such as route selection and frequency of use, based on their satisfaction with EV performance, particularly regarding range and charging convenience. Electric vehicle (EV) users typically report high levels of satisfaction, often accompanied by noticeable changes in their daily routines and driving behaviors. This satisfaction is largely attributed to the perceived advantages of EVs, including their reduced energy consumption and lower emissions [15]. As users gain familiarity with technology, they tend to optimize their driving habits, further enhancing the efficiency and range of their vehicles. The perception and performance of electric vehicle (EV) technology have a substantial impact on travel behavior and usage patterns. Demographic factors such as age, income, and education level influence electric vehicle (EV) adoption and usage trends. Younger individuals, those with higher incomes, and those with higher education levels tend to report greater satisfaction and hold more positive attitudes toward EVs [44]. Notable factors include shifts in driving habits toward greater efficiency, the critical role of reliable charging infrastructure, and high user satisfaction, which are often influenced by demographic factors. These findings emphasize the ongoing need for advancements in EV technology and infrastructure to promote wider adoption and optimize user behavior.

The cost of ownership and usage factors have a positive and significant effect on travel patterns and usage behavior factors (H6). Cost also plays a crucial role in influencing travel patterns and usage behavior. The SEM results show that the cost of ownership and usage positively affect travel behavior, with a path coefficient of β = 0.203 (p < 0.01). Lower operational costs, especially in terms of energy and maintenance, promote the more frequent and longer use of EVs. Driving behavior has a substantial impact on the energy consumption and range of electric vehicles (EVs). Adopting a moderate driving style can lower energy consumption by as much as 30% compared with more aggressive driving habits [44]. In regions where electricity is cheaper and the charging infrastructure is more accessible, EV drivers tend to travel longer distances and use their vehicles more frequently. Conversely, higher costs and limited infrastructure may result in more conservative travel behaviors, such as shorter trips and less frequent charging. Ownership and usage costs, including factors such as income, urban living, travel habits, charging infrastructure, and driving behavior, significantly impact the travel patterns and usage behaviors of electric vehicle (EV) drivers. Electric vehicle (EV) drivers often exhibit distinct travel patterns compared with conventional vehicles. For example, the average and 85th percentile daily trip distances for plug-in hybrid electric vehicles (PHEVs) and hybrid electric vehicles (HEVs) are notably longer than those for conventional vehicles (CVs), with battery electric vehicles (BEVs) following this range [49]. These factors play a critical role in determining EV adoption and economic feasibility in regions, such as Thailand, where urban density and infrastructure development vary. For instance, cities with a growing EV infrastructure, such as Bangkok, can support this transition more effectively. The integration of targeted policies and infrastructure improvements is essential for fostering electric mobility, ensuring that EVs remain a practical and sustainable choice for drivers.

Convenience and accessibility factors have a positive and significant effect on travel patterns and usage behavior factors, with technological perception and performance factors as mediator variables (H7). Mediation analysis demonstrated that technological perception and performance partially mediate the relationship between convenience, accessibility, travel patterns, and usage behavior. The indirect effect was found to be statistically significant, with a coefficient of 0.094 at the 0.01 level. This indicates that improvements in convenience and accessibility positively influence travel patterns and usage behavior through enhanced technological perceptions and performance. Convenience factors, such as the availability of charging stations and the speed of charging, play a vital role in shaping travel behavior and usage patterns of electric vehicles (EVs). Enhancing these aspects can result in increased EV adoption and frequent vehicle use [60]. In essence, as users perceive technology to be more advanced and reliable, they are more likely to travel longer distances and use their EVs more frequently. The driving patterns of electric vehicle (EV) users differ from those of conventional vehicle drivers, with EV users gradually adopting more stable and efficient driving behaviors as they become familiar with their capabilities [31]. The adoption and usage patterns of electric vehicles (EVs) are strongly shaped by factors such as accessibility, convenience, and perceptions of technology. Both the actual and perceived availability of charging infrastructure, along with the perceived benefits and performance of EVs, are key determinants of travel behavior. Improving the convenience and dependability of charging services is essential to encourage greater EV adoption and optimize how drivers use their vehicles.

Convenience and accessibility factors have a positive and significant effect on travel patterns and usage behavior factors, with the cost of ownership and usage factors as mediator variables (H8). Similarly, the cost of ownership and usage also mediates the relationship between convenience, accessibility, travel patterns, and usage behavior. The study found a statistically significant indirect effect, with a coefficient of 0.059 (p < 0.01). This suggests that improvements in convenience and accessibility lead to better travel patterns and usage behaviors by reducing the perceived cost of ownership. As the cost of using and maintaining an EV becomes more manageable owing to increased accessibility, users are likely to use their vehicles more frequently and for longer trips. Convenience and accessibility are key factors that positively influence EV drivers’ adoption and usage patterns of electric vehicle (EV) drivers. The availability of public EV charging facilities has a significant impact on the intention to purchase an electric vehicle. Additionally, perceived and anticipated accessibility to charging stations are important factors, particularly for individuals who do not yet own EV [62]. Lower operating costs and financial incentives such as subsidies enhance the appeal of EVs by acting as mediators. The presence of charging infrastructure, especially fast-charging stations, along with social and locational factors also significantly impacts travel behavior and usage patterns. In summary, a combination of greater accessibility, reduced costs, and supportive policies can effectively promote widespread adoption of electric vehicles.

Wireless charging: Implications for perception and adoption. Wireless EV charging via stationary pads or dynamic in-lane coilscan increase perceived convenience (CA) by removing cable handling and enabling opportunistic top-ups, but it also introduces safety and performance perceptions tied to alignment tolerance and thermal behavior. Recent studies have shown that spatial misalignment can elevate local temperatures and losses, underscoring the need for robust foreign-object/live-object detection, shielding, and thermal management. These safeguards are central to the current reviews of wireless EV charging safety technologies. In parallel, the SAE J2954 framework specifies interoperability and alignment methodologies (e.g., DIPS), which should reduce user-visible failures and improve reliability perceptions over time. Taken together, wireless charging may strengthen CA → usefulness pathways for BEV-oriented users while making perceived risk more salient where standards, detection, or thermal controls are immature. Therefore, we recommend tracking awareness/availability of wireless charging in future data waves and testing moderation by local deployment [63].

6. Conclusions

6.1. Theoretical Contribution

This study offers a significant theoretical contribution by expanding the understanding of electric vehicle (EV) adoption through a multidisciplinary lens. It integrates behavioral economics, the Technology Acceptance Model (TAM), and urban transportation theories to explore the interplay between convenience, accessibility, technological perception, and cost in shaping EV adoption and usage. This study highlights that consumer behavior towards EVs is driven by not only economic rationality but also technological acceptance and environmental considerations. It provides empirical evidence that convenience factors such as the availability of charging infrastructure significantly affect perceptions of technology and influence the cost of ownership. The use of structural equation modeling (SEM) offers a comprehensive framework to examine these complex relationships, contributing to both transportation studies and consumer behavior research by illustrating how these factors collectively influence travel patterns and EV usage behavior, particularly in urbanizing contexts such as Thailand.

6.2. Practical Implication

The findings of this study have several practical implications for policymakers, urban planners, and industry stakeholders aiming to promote EV adoption. This emphasizes the importance of improving accessibility to charging infrastructure, particularly fast-charging stations, and enhancing the convenience of EV ownership, such as home-charging solutions. The study also suggests that reducing the total cost of ownership (TCO) through government incentives and subsidies, combined with improved charging networks, can significantly drive EV adoption. Practical strategies include targeted infrastructure development in urban areas and public awareness campaigns focused on the long-term cost benefits and technological advancements of EVs. Furthermore, the research underlines the need for policies that address socio-demographic factors, as younger and highly educated consumers are more inclined to adopt EVs, whereas overcoming barriers for other groups could lead to broader market penetration.

Policy implications for Thailand and external validity. Our results highlight three levers—Convenience & Accessibility (CA), Technology Perception & Performance (TP), and Cost of Ownership & Usage (CU)—which can be translated into actionable policies.

Thailand-focused actions.

(1) Charging access (CA): Accelerate corridor fast charging on intercity highways and require EV-ready wiring in new condominiums/parking facilities, adopt roaming/interoperability standards to reduce app/connector fragmentation, and prioritize chargers at rail/BRT nodes to support multimodal trips.

(2) Perceptions and performance (TP): national information campaigns with transparent range/charging time benchmarks, mandatory display of battery warranty terms and certified SOH (state-of-health) for used EVs, and public-fleet electrification (taxis, municipal buses) to normalize use and provide visible proof of reliability.

(3) Total cost (CU): time-of-use tariffs with mitigated demand charges for public DC fast charging, low-interest loans or guarantee schemes for batteries, targeted incentives for home/condo chargers, support for a formal second-hand EV market, and battery recycling to lower lifecycle costs and the perceived risk.

Generalizability and China Reference. China’s experience underscores the value of scale, standardized infrastructure, fleet electrification (taxis/ride-hailing/buses), and building code provisions for residential charging. These lessons are directionally consistent with our findings, but policy transfer must be conditional on local constraints: utility regulation and grid capacity, fiscal space for incentives, urbanization patterns, and housing stock (high-rise vs. detached homes). Therefore, we recommend (a) diagnosing baseline conditions (charger density, condo retrofittability, grid headroom), (b) sequencing infrastructure where demand is highest (urban hubs, logistics depots, transit nodes), and (c) testing moderation by EV type and geography (e.g., BEV effects are strongest where public fast-charging density is higher). Under these conditions, our relationships (CA/TP/CU → intention) are likely to generalize to peer developing markets, while the magnitude of the effects will vary with infrastructure maturity and consumer segment mix.

Longitudinal perspective and infrastructure maturity. Although our estimates are cross-sectional, EV adoption is dynamic. As charging networks densify and reliability improves, the salience and magnitude of the key pathways are expected to shift. In particular, the CA → perceived usefulness and CA → intention links should strengthen as access frictions fall; TP (technology perception/performance) may improve through learning and visible fleet use, partially mediating infrastructure effects; and perceived CU (cost of ownership/usage (CU) may decline with tariff design, maintenance experience, and secondary market development. To assess these dynamics, future work should (i) field multi-wave panels and estimate random-intercept cross-lagged or latent growth models to separate within-person changes from between-person differences; (ii) exploit event-study settings (e.g., corridor fast-charger openings, tariff reforms) with time-varying covariates; and (iii) test multi-group trajectories (early adopters vs. late entrants; urban vs. rural). As an interim step with repeated cross-sections, researchers can interact core predictors with a regional charger-density index and calendar time to approximate dynamic moderation. In this way, our current results represent baseline relationships that are likely to intensify for BEV-oriented consumers as networks mature, while the effects for HEV/PHEV segments may plateau.

7. Research Limitations and Future Research Areas

7.1. Limitations of the Research

One limitation of this study is its geographical focus on Thailand, which may limit the generalizability of the findings to other countries with different levels of infrastructure development and consumer behavior. This study predominantly captures the behavior of urban consumers in a rapidly urbanizing environment, but rural areas with less developed charging infrastructure are not fully represented. Additionally, reliance on self-reported data from surveys could introduce biases, as respondents might overstate their positive attitudes toward electric vehicles (EVs) or underreport the barriers they face in adoption. The study also focuses more heavily on factors such as accessibility and cost, potentially overlooking the influence of psychological factors such as brand loyalty or cultural attitudes toward green technologies. Furthermore, the cross-sectional nature of the study means that it captures consumer behavior at a single point in time, limiting its ability to assess how EV adoption behavior evolves as technology and infrastructure improve over time.

7.2. Suggestions for Further Research

Future research should expand this study by examining electric vehicle (EV) adoption across diverse geographic regions, including rural areas and countries with different levels of infrastructure maturity. Comparative studies between countries with established EV markets and emerging markets offer valuable insights into how infrastructure, policy, and consumer behavior interact globally. Longitudinal studies would also be beneficial for tracking how EV adoption behaviors change over time, particularly as new technologies (e.g., faster charging systems and improved battery life) become more widespread. Researchers should consider psychological and cultural factors such as environmental identity, green consumerism, and resistance to change to deepen the understanding of the motivators and barriers to EV adoption. Additionally, future research could explore how different socioeconomic groups respond to government incentives and policies designed to promote EV adoption, thereby helping policymakers design more targeted and effective interventions. Finally, investigating the role of public awareness campaigns and how communication strategies influence both perceived convenience and technology perception would provide practical insights for industries that aim to boost EV adoption.