Research on factors influencing users’ intentions to use battery swap technology: a comprehensive model analysis Authors: Xing Xu, Siyu Feng, Zhiling He, Tiansheng Xia Published: 2026 ABSTRACT AbstractThis study develops a user intention model for the battery-swapping mode based on the Theory of Planned Behavior (TPB) and the Unified Theory of Acceptance and Use of Technology 2 (UTAUT2), integrating perceived risk as a key antecedent variable. A total of 516 participants were recruited in this study, and data on performance expectancy, effort expectancy, hedonic motivation, price value, habits, facilitating conditions, perceived risk, attitude, subject norm, perceived behavioral control, and willingness to use battery swapping technology were collected via questionnaires. The results show that attitude, subjective norms, and perceived behavioral control exert significant direct effects on users’ intention to adopt the battery-swapping mode. Performance expectancy, effort expectancy, hedonic motivation, and facilitating conditions influence intention indirectly through attitude. By establishing a more granular model, this study provides a clearer understanding of user acceptance of automotive battery-swapping technology. The findings offer an evidence-based approach for examining user intention and provide actionable implications for enterprises and policymakers seeking to refine battery-swapping strategies and support the development of the electric vehicle sector.Keywords: Battery swap technology, Electric vehicle, TPB model, UTAUT2 model, Perceived riskSubject terms: Business and management, Business and management, Information systems and information technology, Psychology, Psychology Introduction As an innovative solution to environmental problems and energy crises, electric vehicles (EVs) are leading the development of the automotive industry. In order to promote the popularity of electric vehicles, the Chinese government has adopted a number of incentive measures in attempt to attract more consumers to use them. However, electric vehicles have experienced limited success compared to existing vehicles and their popularity1. Consumers’ concerns about the range and charging time of electric vehicles are the main reasons that influence their purchase and use decisions, many consumers believe that these aspects of EVs cannot meet their needs2. Therefore, EVs’ limited cruising range restricts consumers’ use of these vehicles3. Battery swap technology (BST) is one important method used to solve the problem of the limited cruising range of electric vehicles. Different from the traditional plug-in charging method, battery swap technology can quickly replace a depleted battery with a fully charged battery at a battery swap station while the electric vehicle is driving; this process takes about 5 min, thereby extending the driving range of electric vehicles4. This method can significantly shorten the charging time of electric vehicles and improve their efficiency5. The introduction of BST could solve the problems of “range anxiety” and “charging waiting anxiety” faced by consumers when using electric vehicles. Moreover, this technology could also extend the service life of a battery and increase its value throughout its life cycle6. BST has gradually been implemented into and promoted by the automobile manufacturing industry, and increasing numbers of companies have begun to join this field7. Overall, BST demonstrates considerable potential, and its development within the electric vehicle sector shows a positive and promising trajectory. However, although BST solves the main cruising range problem of electric vehicles, there is still much room for improvement in terms of the popularity of BST. Currently, research on BST focuses on the construction of battery swap stations, battery swap service strategies, and research on battery technology8–10. However, there are relatively few studies focused on user acceptance of BST. Chen et al. compared users’ preference choices between charging and battery swapping, identifying the key factors that shape their selection of energy-replenishment methods11. Because users’ acceptance and attitudes toward battery-swapping technology (BST) are crucial to its widespread adoption7, developing appropriate strategies aimed at reducing user concerns and barriers to use is essential for promoting broader uptake of BST. In current theoretical research on technology acceptance, major theoretical models such as the theory of planned behavior (TPB) and the unified theory of acceptance and use of technology (UTAUT2) are widely studied. The TPB theoretical framework suggests that behavior related to interest in using a product can be predicted by attitude (ATT), subject norm (SBN), and perceived behavioral control (PBC)12. The UTAUT theoretical framework, initially proposed by Venkatesh et al.13, was designed to integrate performance expectancy and social influences as key variables capable of influencing behavioral interest in the adoption of new technology, along with the associated facilitating conditions. The UTAUT model was subsequently developed into UTAUT2 by adding other variables, such as hedonic motivation, price value, and habits, which are believed to influence interest in using the technology and can predict the eventual use of the new technology in the future14. A previous study conducted by Yuen et al. developed a research model that integrates the UTAUT2 and TPB models to accurately predict technology usage behavior15. In a more recent study, Gunawan et al. further expanded this framework by including perceived risk in addition to the TPB and UTAUT2 models16, enabling the effective forecasting of an individual’s intention to adopt electric vehicles. Zhang et al. employed the UTAUT2 model to analyze users’ technology acceptance of battery-swapping services in micro-mobility contexts17. Chanda et al. applied the TAM and TPB frameworks to examine consumers’ intentions to use battery-swapping technology (BST)18. Nguyen-Phuoc et al., focusing on the adoption of electric motorcycles, developed an integrated model based on TPB and UTAUT2 to predict users’ intention to adopt electric motorcycles19.These studies demonstrate that the TPB model, comprising the variables ATT, SBN, and PBC, significantly influences the behavior of technology usage. At the same time, the UTAUT2 model includes performance expectancy (PE), effort expectancy (EE), hedonic motivation (HM), price value (PV), habits (HBs), and facilitating conditions (FCs). These factors can be used to measure the influence of individuals on technology use behavior. In these studies, we also found that the perceived risk (PV) factor is capable of influencing an individual’s behavior towards technology use. Overall, the Theory of Planned Behavior (TPB) provides a socio-psychological foundation for understanding the formation of behavioral intention, while UTAUT2 supplements this framework by introducing technology-related determinants of user behavior. The integration of these two theories enables a more comprehensive explanation of how users develop attitudes and make decisions when adopting new technologies. Accordingly, these theoretical models have demonstrated strong predictive power in identifying key variables influencing users’ technology adoption behaviors, which offers a clear direction for the present study. This research builds upon previous work by Yuen et al.15, who confirmed the applicability of TPB and UTAUT2 in predicting users’ intention to adopt battery-swapping technology (BST). Although existing integrated models provide meaningful insights into technology adoption, they remain somewhat fragmented and lack a unified framework that incorporates TPB, UTAUT2, and perceived risk. While some studies have included perceived risk, they have not thoroughly examined its role within the combined UTAUT2–TPB structure. To address this gap, the present study proposes a more comprehensive model. The objectives of this study are: (1) to develop a user intention model for battery-swapping adoption that integrates TPB, UTAUT2, and perceived risk; (2) to examine the factors influencing users’ acceptance of BST; and (3) to provide practical recommendations for the automotive industry based on the findings. The structure of this article is as follows: the subsequent section introduces the theoretical foundation of the study, followed by the development of research hypotheses and the conceptual model. The next section presents detailed explanations of each hypothesized construct. Subsequently, structural equation modeling is employed to analyze the empirical data and report the results. The final section discusses the theoretical and practical implications, outlines the limitations of the study, and proposes directions for future research. Literature review and hypothesis Literature review on battery swap technology Multiple studies focused on BST have been conducted. For example, some studies have focused on collaborative switching service management schemes based on battery heterogeneity8, as well as addressing battery degradation issues and formulating issues related to battery usage and charging9, among other pertinent topics. These studies mainly focus on the construction of battery swap stations, battery swap service strategies, and battery technology research. However, a limited number of studies have investigated users’ intentions to adopt BST. Existing studies on users’ intention to adopt BST indicate that several factors can significantly promote adoption, including perceived usefulness, attitude, subjective norms, knowledge, convenience, as well as hedonic and aesthetic qualities4,7,20,21. At the same time, prior research has shown that perceived risk, safety concerns, price sensitivity, and technology anxiety may hinder users’ intention to use BST4,7,17. Although the existing literature has identified a variety of determinants of BST usage intention, the findings remain somewhat fragmented and lack a systematically integrated analytical framework. To address this gap, this study extends the TPB-UTAUT2 framework by incorporating perceived risk, thereby proposing a more comprehensive model to systematically explain and predict users’ intention to adopt BST. Theories, models, and hypotheses In this study, we developed a BST user intention model by integrating the TPB and the UTAUT2 to predict users’ intentions to employ BST, and we also investigated the factors that could influence users’ attitudes towards utilizing BST. Amidst a multitude of conceptual frameworks, the TPB model, which is constituted by attitude, subject norm, and perceived behavioral control, has been employed to directly forecast intentions to adopt battery swap technology12. In the UTAUT2 theoretical model, PE, EE, HM, HBs, and PV were found to impact the degree of adoption, while attitude influenced the propensity to utilize BST and facilitating conditions influenced individuals’ perceived behavioral control15. In a recent investigation of the determinants of consumers’ adoption of pure electric vehicles22, the presence of facilitating conditions was found to have a direct influence on the propensity to utilize BST. Therefore, drawing on UTAUT2, this study employs performance expectancy, effort expectancy, hedonic motivation, habit, price value, and facilitating conditions as components of the UTAUT2 model to examine the determinants of users’ attitudes toward using BST. Attitude is conceptualized as a second-order construct reflected by these variables, allowing for a more comprehensive representation of users’ intentions in real-world contexts. Perceived risk plays an important role in shaping individual usage behavior23. In the context of battery swapping, users may be concerned about risks associated with battery safety, lifespan, and other uncertainties; thus, perceived risk is incorporated into the model to explain potential negative effects on user attitude and intention. Our research presents an integrated model of user attitude and intention to use BST, as illustrated in Fig. 1. Research model. This study integrates the TPB and UTAUT2 models. TPB is applied to predict users’ intention to use BST24, whereas UTAUT2 further explores behavioral patterns related to the adoption and use of new technologies14. TPB posits that attitude, subjective norms, and perceived behavioral control positively influence users’ intention to adopt BST. UTAUT2 variables—performance expectancy (PE), effort expectancy (EE), hedonic motivation (HM), price value (PV), habit (HB), and facilitating conditions (FC)—are hypothesized to positively affect users’ attitudes toward BST, thereby influencing their eventual adoption behavior16. In contrast, perceived risk is a major barrier that negatively affects consumers’ attitudes toward BST25. Accordingly, this study proposes a TPB–UTAUT2 attitude model that offers a more precise explanation of the mechanisms underlying users’ acceptance of BST. The determinants of intention to use BST based on TPB The TPB can predict various intentions and behaviors26. According to research by Icek Ajzen12, a person’s behavior is determined by their behavioral intention, which is influenced by their attitude towards a particular behavior, subjective norms, and perceived behavioral control. This theoretical construct can be used to predict consumers’ intentions to use the technology24. Therefore, the TPB theoretical model was employed in this study to forecast users’ intentions to adopt BST. Attitude The initial determinant of the intention to employ BST is attitude, which pertains to an individual’s positive or negative appraisal of the behavior12. Furthermore, attitude serves as a crucial predictive factor of behavioral intention27,28. This means that the higher a user’s positive evaluation of a behavior is, the stronger their impulse to intend to use is29. If a user responds positively to a product or service, their likelihood of using it is higher when they experience feelings of liking, comfort, or pleasure13. This suggests that when consumers rate their attitudes towards BST as higher, their intentions to use BST become stronger. Therefore, the hypothesis that we formulated in this regard is as follows: H1 Attitude positively influences consumers’ intentions to use BST. Subject norm The second determinant of the propensity to employ BST pertains to subject norm, which refers to the societal external pressure imposed on individuals to undertake the behavior in question12. This pressure emanates from familial, peer, and colleague support, as well as media exposure, among other sources28. A higher degree of external pressure increases the likelihood of adopting a particular behavior30. This implies that when users observe influential friends, relatives, or authoritative experts utilizing BST, they might feel compelled by external factors, consequently fostering an intention to employ BST4. Consequently, the hypothesis that we posited is as follows: H2: Subject norm positively influences consumers’ intentions to use BST. Perceived behavioral control The third variable influencing the propensity to adopt BST is perceived behavioral control, which refers to an individual’s personal feeling and subjective evaluation of control ability and efficiency when achieving goals or performing behaviors12,31. This indicates that an individual’s perceived ability to control such behaviors is elevated when they have confidence in their abilities, the available resources, and self-efficacy32. In this research, users who possess adequate BST information and resources and believe that utilizing BST is straightforward are more likely to intend to employ it4. Consequently, the formulated hypotheses are as follows: H3 Perceived behavioral control positively influences consumers’ intentions to use BST. Factors influencing attitude towards the use of BST based on the UTAUT2 model The UTAUT2 theoretical model, which is based on the six variables of PE, EE, HM, PV, HBs, and FCs, is employed to explain users’ behavioral intention to adopt and utilize new technologies13. Past studies have combined the UTAUT2 and TPB models to forecast interest in employing this technology. Furthermore, the UTAUT2 model has been utilized to predict users’ attitudes towards new technologies15,16. Thus, this research employs the UTAUT2 model to investigate the determinants influencing users’ attitudes towards the utilization of BST. Performance expectancy The initial factor believed to shape attitudes towards the utilization of BST is performance expectancy. PE is defined as “the degree to which the implementation of a technology yields benefits for consumers while engaging in specific activities.” This implies that an individual’s preference for or aversion towards an object is contingent with their assessment of its performance. In Gunawan et al.‘s research14, a significant influence of performance expectancy on attitudes was observed. Moreover, PE is also deemed to be the primary determinant of users’ propensity to adopt new technologies15. BST offers multiple advantages and utilities for users in their daily driving routines. For instance, BST can abbreviate users’ charging time, mitigate charging apprehensions, and enhance the efficiency of electric vehicles4. Additionally, BST can prolong the lifespan of batteries and confer environmental friendliness by implementing centralized battery management, recycling techniques, and other operational methods. These sustainable strategies in a closed-loop configuration can generate profitable outcomes while mitigating potential environmental pollution risks triggered by battery disposal6, among others. These benefits hold the potential to sway users’ attitudes towards BST positively. If users perceive the utilization of BST as advantageous for themselves, they will develop a favorable attitude towards it, accordingly having an influence on their intention to employ BST. Therefore, the following hypothesis was formed: H4 Performance expectancy positively influences consumers’ attitudes toward BST. Effort expectancy The second factor identified as influencing attitudes towards the use of BST is effort expectancy, which reflects an individual’s perceived ease of using the system13. As the UTAUT2 theory was derived from the TAM, the concept of EE is analogous to the “perceived ease of use” in the TAM, which pertains to the degree to which a consumer anticipates that using the system will be simple and uncomplicated14. The ease of using a technology significantly impacts an individual’s attitude towards it33. This implies that individuals are more likely to adopt and utilize the technology if they perceive it to be relatively straightforward to use. In our research, effort expectancy represents users’ perceived ease of using BST34. And users will possess a positive attitude towards using BST if they consider it to be a relatively easy technology for them to learn and employ. Therefore, the following hypothesis was developed: H5 Effort expectancy positively influences consumers’ attitudes towards BST. Hedonic motivation The third factor attributed to shaping attitudes towards the use of BST is hedonic motivation, defined as “the pleasure or happiness experienced by users when employing a technology”13. Factors such as satisfaction, pride, emotions, and other subjective feelings generate hedonic motivation35. Individuals who perceive technology use as a source of happiness are more likely to be attached to it13. This indicates that individuals with higher hedonic motivation exhibit a more favorable evaluation of technology use15,36. HM serves as a crucial determinant of attitudes towards use16. In Ajzen’s study37, it was discovered that during the initial stages of adopting a novel technological product, users were more likely to find novelty appealing. The novelty and freshness provided by BST may elicit pleasure and enjoyment, leading users to believe that using BST can bring them pleasure, entertainment, or fulfillment, accordingly fostering a positive attitude towards it. Therefore, the hypothesis formed in this regard is as follows: H6 Hedonic motivation positively influences consumers’ attitudes towards BST. Price value The fourth factor pertains to the influence of price value on attitudes towards BST from the consumer’s perspective38. PV is characterized as “the trade-off between the economic costs and benefits of a technology for consumers”. When the benefits derived from the technology exceed individuals’ expectations, the product value potentially registers as higher for consumers13. The notion that a positive price value arises when consumers perceive greater advantages and utility derived from a product or service than the monetary cost incurred. Correspondingly, individuals with heightened awareness of the technology’s value are more inclined to pay for its usage15. Empirical evidence from previous studies supports the positive impact of price value on consumers’ attitudes towards utilization16. In the current investigation, price value is associated with the economic and temporal costs of BST. Assuming users perceive that employing BST can confer advantages in terms of economy and time, and its value surpasses their personal expectations, their attitudes towards utilizing BST are likely to be more favorable. Therefore, the hypotheses formed are as follows: H7 Price value positively influences consumers’ attitudes towards BST. Habit The fifth factor attributed to shaping attitudes towards the use of BST is habit, which is characterized as “an individual’s automated, unconscious behavioral responses resulting from past experiences in a specific context”13. While Kim et al. defined habit as past usage, it can serve as a predictive factor for determining whether particular technology will be employed in the future39. Previous research has employed various proxies to measure habit, such as past behaviors, reflexes, and personal experiences40,41. Past behaviors are regarded as antecedent actions and serve as a proxy measure. Reflexes encompass sequences of innate behaviors that are unlearned. Personal experience refers to the user’s familiarity with technology, derived from stable routines, norms, and habits. Furthermore, habitual prior usage enhances interaction and familiarity with novel technology, impacting consumers’ future adoption of the technology13. This indicates that if users develop a habit of employing BST, it will be encoded in their consciousness, resulting in an unconscious positive attitude towards the utilization of BST, which is consistent with the findings of previous research15. Therefore, the hypothesis formed in this regard is as follows: H8 Habits positively influence consumers’ attitudes towards BST. Facilitating conditions The sixth measure investigates the influence of facilitating conditions on attitudes towards BST, which refers to “the support and resources from the external environment encountered by individuals when utilizing a new technology”13. FCs are believed to impact perceived behavioral control, suggesting that the external environment’s support and resources influence an individual’s ability to manipulate and utilize technology effectively16. However, the integration of the UTAUT2 model with the TPB theoretical model in this study highlights the examination of the effects of specific factors on consumer attitudes towards technology. Cognitive and affective responses to technology use are influenced by facilitating conditions, which exert a positive influence on individuals’ attitudes towards utilizing technology when perceived as positive, beneficial, or helpful. Facilitating conditions have the potential to enhance user experience, mitigate potential barriers, and augment comfort levels when utilizing technology or services42. This study aims to investigate the extent to which external facilitating conditions influence users’ perceptions of BST, their propensity to develop positive attitudes towards it, and whether such positive attitudes translate into actual intentions to employ BST. Therefore, the following hypothesis was developed: H9 Facilitating conditions positively influence consumers’ attitudes toward BST. Perceived risk determinants of ATT on BST Perceived risk refers to the unexpected outcomes that consumers may encounter when using a product or service, resulting in dissatisfaction43. This perceived risk can manifest in various forms, including financial, performance, social, psychological, and physical hazards. Consequently, if potential users exhibit skepticism towards BST, they might refuse to use it. The impact of perceived risk on deterring people from adopting BST has been significantly emphasized by Wang et al.44. In this study, potential users of BST may harbor concerns regarding the utilization of swapped batteries and dread potential combustion events due to safety risks and economic losses. These apprehensions could potentially influence their social standing and self-image, thereby serving as barriers to the adoption of BST4. It is contended that BST is a cost-intensive technology necessitating substantial initial investments, behavioral modifications, and potential upgrades to domestic infrastructure. These factors engender certain purchase risks, including potential adverse effects, hazards, irritations, and financial losses. Consequently, they might give rise to negative attitudes towards BST and reduce users’ propensity to use it45. Perceived high-risk situations render users more susceptible to developing negative attitudes and intentions towards the use of BST. Therefore, the hypothesis formed in this regard is as follows: H10 Perceived risk negatively influences consumers’ attitudes towards BST. Methodology Measurement tools According to the relevant maturity scale, combined with the actual situation of BST, the relevant measurement items in the questionnaire were appropriately adjustment. We set 11 variables for the user’s BST usage intention model and measured these items using a 5-point Likert scale, where 1 denoted “strongly disagree” and 5 represented “strongly agree”. The questionnaire was adapted from previous studies in the area of technology acceptance4,7,16. Details of the questionnaire can be found in Appendix A. Data collection and sample characteristics The target population of this study comprises individuals who have used or possess a basic understanding of electric vehicles (EVs). The sample focuses on early adopters of BST46, who are predominantly younger users and thus well-suited for identifying key factors influencing the initial adoption of BST. Future studies may expand to broader populations to validate the model’s generalizability. A convenience sampling approach was employed to facilitate rapid data collection and implementation, which is effective when surveying large groups. China was selected as the study region because its EV industry has received strong governmental support and exhibits a highly active market47. Data were collected through an online questionnaire administered via the “Question Star” survey platform. Prior to the formal survey, a pilot test involving 60 participants was conducted to assess the questionnaire’s validity. The pilot results indicated that the instrument was suitable for large-scale data collection. The questionnaire design was based on established measurement scales and was adapted to the context of battery-swapping services to ensure cultural relevance and accurate measurement of the target constructs. After minor revisions based on pilot feedback, the formal survey was distributed to the general public, maintaining demographic consistency with the pilot sample. Participants completed the survey after being informed of its confidentiality. The sample’s demographic characteristics are presented in Table 1. A total of 590 questionnaires were collected for this study. After excluding invalid responses due to short completion times, obvious inconsistencies, or consistently selecting the same number of options, 516 valid questionnaires were ultimately obtained, yielding a valid recall rate of 87.4%. Among the valid respondents, 51.9% were male and 48.1% female. The majority were aged 18–30, a demographic widely recognized as early adopters of new energy vehicles and one that is highly familiar with digital technologies, facilitating the collection of reliable data. Additionally, 50.8% of participants reported strong interest in EVs, and 36% expressed an intention to purchase an EV within the next two years. Table 1Description of the distribution of sample characteristics.VariableOptionAmountPercentageGenderMale26851.9Female24848.1Age18ཞ3018034.931ཞ4017133.141ཞ5013325.8EducationHigh school or below5210.1Junior College15229.5Bachelor’s Degree20740.1Master’s Degree8817.1Doctorate or Higher173.3Marital statusMarried with children19437.6Married without children12724.6Unmarried16832.6Divorced275.2Monthly incomeLess than 5000rmb155305001ཞ10000rmb16431.810001ཞ15000rmb11221.7Over 15001rmb8516.5New energy vehicleslevel of interestVery interested13826.7Moderately interested26250.8Not interested but willing to learn more8917.2Not interested at all275.2New energy vehiclespurchase plansPlan to purchase within the next six months9418.2Plan to purchase within the next year16031Plan to purchase within the next two years18636No plans for now, but may purchase in the future5610.9No plans, but would like to learn more203.9 Description of the distribution of sample characteristics. New energy vehicles level of interest New energy vehicles purchase plans The descriptive statistics further showed that the overall sample had a standard deviation of 23.550, a skewness of −1.663, and a kurtosis of 2.423. These values fell within acceptable ranges, indicating that the data did not exhibit serious deviations from normality. The study protocol received approval from the Research Ethics Committee of Guangdong University of Technology. All procedures were performed in compliance with the ethical standards outlined in the Declaration of Helsinki (1964) and its subsequent amendments. Prior to their involvement, written informed consent was secured from every participant, each of whom was advised of their unconditional right to withdraw at any time. Experimental data analyses Reliability and validity analyses Results of the reliability analysis An initial evaluation of the scales’ validity and reliability was conducted using the collected data, with Cronbach’s alpha analysis being employed to assess the reliability of the current study. As illustrated in Table 2 , which presents the results of the reliability analysis, the Cronbach’s coefficient values for both the overall scale and each dimension exceeded the recommended benchmark value of 0.70, ranging from 0.8 to 148. This finding indicates that the questionnaire demonstrates a high level of reliability. Table 2Scales for reliability and convergent validity.ConstructItemLoadingαAVECRATTATT10.8030.8660.6180.866ATT20.780ATT30.772ATT40.789SBNSBN10.7770.8130.5930.814SBN20.754SBN30.779PBCPBC10.7520.8130.5770.804PBC20.771PBC30.779ITUITU10.7330.8030.5850.876ITU20.743ITU30.796ITU40.773ITU50.777PEPE10.7820.8190.5320.819PE20.676PE30.737PE40.719EEEE10.7570.8410.5720.842EE20.771EE30.747EE40.750HMHM10.8010.8560.5980.856HM20.776HM30.772HM40.742PVPV10.7500.7840.5490.785PV20.724PV30.748HBsHB10.8020.8600.6060.860HB20.777HB30.760HB40.775FCsFC10.7620.8350.5600.836FC20.771FC30.721FC40.739PRPR10.7660.8560.5980.856PR20.802PR30.769PR40.755Note: α = Cronbach’s alpha; CR = composite reliability; AVE = average variance extracted; ATT = attitude; SBN = subject norm; PBC = perceived behavioral control; ITU = intention to use BST; PE = performance expectancy; EE = effort expectancy; HM = hedonic motivation; PV = price value; HBs = habits; FCs = facilitating conditions; PR = perceived risk. Scales for reliability and convergent validity. Note: α = Cronbach’s alpha; CR = composite reliability; AVE = average variance extracted; ATT = attitude; SBN = subject norm; PBC = perceived behavioral control; ITU = intention to use BST; PE = performance expectancy; EE = effort expectancy; HM = hedonic motivation; PV = price value; HBs = habits; FCs = facilitating conditions; PR = perceived risk. Results of the validity analysis The results of the model fitness test in this study indicate that the model exhibits a CMIN of 1.414, which is within the acceptable range of 1–3, and an RMSEA of 0.038, which falls within the excellent range (< 0.05). Furthermore, IFI, TLI, and CFI all achieve excellent levels, exceeding 0.949. Consequently, integrating the outcomes of this analysis suggests that the conceptual model constructed in this study exhibits a good fit as a measurement model. Under the precondition that the confirmatory factor analysis (CFA) model has a good fit, the average variance extracted (AVE) and composite reliability (CR) of each dimension within the scale were further evaluated. Based on the results presented in Table 2, it can be discerned that all the dimensions achieved an AVE value surpassing 0.5 and a CR value exceeding 0.7, suggesting satisfactory convergent validity and composite reliability. Simultaneously, this study examined the differential validity of each dimension. While convergent validity and combined reliability analyses focus on measuring quality within a single dimension, differential validity analyses concentrate on discerning differences between different dimensions. Based on the results presented in Table 3, the standardized correlation coefficients among the dimensions in this differential validity test are lower than the square root of their corresponding AVE values, indicating good differential validity between these dimensions. Table 3Differential validity test results of each dimension.VariablePEEEFCsHMPVHBsPRATTSBNPBCITUPE0.532EE0.3110.572FCs0.3700.2800.560HM0.2880.3710.3110.598PV0.3770.3520.3900.3590.549HBs0.3770.3160.3610.2770.2900.606PR0.3600.4270.4250.3330.3250.2780.598ATT0.3650.4150.3640.3780.2930.2610.4170.618SBN0.3420.3070.3120.2440.2890.3260.2820.2870.593PBC0.4860.3390.3610.2520.3750.3130.3330.3590.3390.577ITU0.5530.5120.5590.4600.5010.5600.5330.4640.4920.5090.585Note: Scores in bold represent the square root of average variance extracted for a construct. Differential validity test results of each dimension. Note: Scores in bold represent the square root of average variance extracted for a construct. Results of exploratory factor analysis First, KMO and Bartlett’s sphericity tests were conducted to assess the suitability of the collected data. As shown in Table 4, the KMO value for the scale was 0.817, exceeding the recommended threshold of 0.70 proposed by Kaiser50. This indicates that the sample data is well-suited for factor analysis. Concurrently, Bartlett’s sphericity test yielded χ² = 2077.579, p < 0.001, reaching statistical significance. This indicates the correlation matrix is non-identity and suitable for factor analysis. Table 4KMO and Bartlett’s Test.KMO Measure of sampling adequacy0.817Bartlett’s Sphericity TestApproximate Chi-Square2077.579Degrees of Freedom45Significance0 KMO and Bartlett’s Test. Based on these findings, factor extraction employed Principal Component Analysis (PCA) with Varimax rotation. Results showed cumulative explained variance of 72.023%, exceeding the recommended threshold of 50%51, indicating that extracted factors adequately reflected the information of original variables. All measurement items exhibited factor loadings exceeding 0.6 on their respective factors, demonstrating strong convergent validity of the factor structure. The scale structure demonstrated good fit and construct validity in the exploratory factor analysis, validating the rationality of the designed scale structure. Structural model and hypotheses test We employed SEM analysis to assess the fit of the data to the structural model and its relationship with the hypotheses. Confirmatory factor analysis (CFA) is suitable for theory-driven model validation, enabling simultaneous evaluation of the fit of measurement models and structural models while providing accurate parameter estimates and hypothesis testing results52. The analysis was conducted using AMOS 26.0 software. The outcomes of the SEM analysis demonstrate a good concurrence between the data and assumptions of the structural model. The fitted values are as follows: CMINCF = 2.407, RMSEA = 0.052, IFI = 0.884, TLI = 0.877, and CFI = 0.884. These values align with a well-fitting structural model, and the recommended benchmark values corroborate this conclusion. The evaluation of the proposed structural paths’ relationship is based on their statistical significance level and standardized loading (refer to Fig. 2). Path coefficients represent the associations between variables; the magnitude of the path coefficient effectively indicates the strength of the relationship between different variables, with positive values indicating a positive correlation and negative values indicating a negative correlation. Table 5 summarizes the outcomes of the structural model and the hypothesized relationships among the variables. To examine common method variance, this study employed Harman’s single-factor test. Results indicated that the unrotated factor analysis extracted 11 factors, with the largest factor explaining 26.911% of the variance—below the 40% threshold53. Therefore, no significant common method variance was present in the study data. Structural model results. Note: *** p < 0.001; n/s = not significant. Table 5Scales for reliability and convergent validity.Dependent variableHypothesisPathEstimateC.R. P HypothesissupportedATTH4PE→ATT0.1863.921***SupportedH5EE→ATT0.2495.247***SupportedH9FCs→ATT0.1763.738***SupportedH6HM→ATT0.2164.581***SupportedH7PV→ATT0.0370.7480.455Not supportedH8HBs→ATT0.0481.0330.302Not supportedH10PR→ATT−0.232−4.914***SupportedITUH1ATT→ITU0.3527.137***SupportedH2SBN→ITU0.3426.996***SupportedH3PBC→ITU0.3316.751***SupportedNote: *** p < 0.001; n/s = not significant. Scales for reliability and convergent validity. Note: *** p < 0.001; n/s = not significant. This study employed regression analysis in SPSS to obtain R² values for assessing the explanatory power of endogenous variables. Results indicate that the R² for Attitude (ATT) was 0.256, suggesting the model explains approximately 25.6% of its variance; the R² for Intention to Use (ITU) was 0.337, indicating the model explains approximately 33.7% of its variance. Both demonstrate good explanatory power. Simultaneously, multiple collinearity testing was conducted on the independent variables in SPSS, extracting the variance inflation factor (VIF). Results showed that the VIF values for variables predicting ATT ranged from 1.230 to 1.332, while those predicting ITU ranged from 1.117 to 1.155. Both ranges were significantly below the commonly used threshold of 551. Therefore, multicollinearity was not an issue in this study, and the structural path estimates were reliable. To examine the model’s stability across different gender groups, this study conducted a multi-group structural equation modeling (SEM) analysis. As shown in Tables 5, 6 the unconstrained model demonstrated good fit (χ² = 2909.636, df = 1630, CMIN/DF = 1.785, RMSEA = 0.039, CFI = 0.871) and served as the baseline model. Subsequently, measurement weights, structural weights, structural residuals, and measurement residuals were sequentially added as constraints. The fit indices remained stable, with minimal changes in CFI/TLI and RMSEA < 0.04. Further nested model comparisons revealed that the incremental χ² values for each constrained model relative to the unconstrained model were non-significant (p = 0.899–0.295), indicating that adding constraints did not significantly degrade model fit. This supports metric invariance and structural invariance. The models demonstrated robust stability across gender groups, providing a reliable foundation for subsequent cross-group comparisons of path coefficients. Table 6Gender-Specific Multigroup Invariance Test.MODLECMINDFCMIN/DFΔχ²ΔDF P CFITLIRMSEAUnconstrained2909.63616301.785–––0.8710.8640.039Measurement weights2933.64316641.76324.007340.8990.8720.8680.039Structural weights2944.76316741.75935.127440.8280.8720.8680.038Structural residuals2948.75716761.75939.121460.7540.8720.8680.038Measurement residuals3004.27217181.74994.636880.2950.8710.870.038 Gender-Specific Multigroup Invariance Test. The structural model results show that most of the hypotheses are verified, and the collected user data highly supports our hypothesized model. Specifically, in the user BST usage intention model we proposed, two aspects were studied. On the one hand, the impact of the UTAUT2 on attitudes towards BST was studied. This research showed that PE (β = 0.16, p < 0.001), EE (β = 0.21, p < 0.001), HM (β = 0.18, p < 0.001), and FCs all have a positive impact on attitudes towards BST; however, perceived risk has a negative impact on attitudes towards BST (β = −0.20, p < 0.001). However, PV and HBs have no impact on attitudes towards BST (β_PV = 0.03, p = 0.455; β_HB = 0.04, p = 0.302). On the other hand, we have also studied the impact of the TPB on intention to use BST. The research results show that ATT influenced by the UTAUT2 model significantly affects consumers’ intentions to use BST (β = 0.29, p < 0.001). At the same time, SBN and PBC also have a positive impact on users’ intentions to use BST (βSBN = 0.24, p < 0.001/βPBC = 0.23, p < 0.001). Within the UTAUT2 components influencing attitudes toward BST, performance expectancy, effort expectancy, hedonic motivation, and facilitating conditions all showed significant positive effects, whereas perceived risk demonstrated a significant negative effect. In contrast, price value and habit did not exhibit significant relationships with attitudes. Regarding the TPB pathways, attitude—shaped by UTAUT2 variables—significantly predicted intention to use BST, alongside subjective norms and perceived behavioral control, both of which exerted positive effects. Collectively, these findings confirm that the integrated UTAUT2–TPB model provides a robust explanation of users’ attitudes and intentions toward adopting BST. Discussion and implications Discussion of the results Based on the TPB and UTAUT2, in this study, we have constructed a model of user willingness to use BST by considering perceived risk. This model explores the factors that affect users’ willingness to use BST and obtains meaningful results. The results of BST utilization intentions First, ATT significantly influences consumers’ intentions to use BST. This finding aligns with that of previous studies4,16,28, suggesting that consumer attitude is a prerequisite for the intention to use BST. Adu-Gyamfi et al. stated that the better consumers perceive the benefits gained from using BST to be, the greater their motivation to intend to use the technology is7. Therefore, firms can enhance consumers’ positive attitudes towards BST, which will stimulate more users to try and utilize it. Second, SBN also affects consumers’ intentions to use BST, a result which is consistent with previous studies4,16. This finding implies that in the context of consumers’ decision-making process regarding the adoption of BST, an increase in external pressure to implement BST in their environment fosters greater interest in utilizing BST in the future4,16. Consequently, both companies and governments should provide suitable incentives to motivate more consumers to experiment with and adopt BST, such as exerting social pressure on those who employ BST. Thirdly, PBC significantly influences consumers’ intentions to use BST, which is consistent with the findings of previous studies15,16. This finding suggests that stronger consumer beliefs regarding BST lead to a heightened propensity to adopt it. To augment the base of BST users, firms should prioritize boosting users’ confidence and mitigating their concerns, thereby strengthening users’ trust and certainty in effectively leveraging BST. The results of the factors influencing attitudes toward BST use The findings from the BST usage intention survey reveal that consumer attitudes play a decisive role in shaping their adoption of BST. Consequently, electric vehicle enterprises should strive to foster favorable user attitudes towards BST. Drawing upon the results of this research, companies may concentrate on the following aspects and provide corresponding incentives. First, the role of PE in explaining consumers’ adoption of new technologies is crucial15,16. This implies that consumers’ attitudes towards BST become more positive when they anticipate gaining significant performance improvements and benefits from its usage. Consequently, companies can enhance users’ performance expectations of BST by expanding the number of switching stations, providing more convenient charging methods, and offering rapid charging services. Second, EE has a positive effect on attitudes toward using BST. If individuals believe that the process of using a new technology will be relatively simple and easy, they are more likely to choose to use it15,16. This suggests that users’ attitudes towards using BST are more positive if they believe that the process of using BST will be relatively easy. Therefore, companies can provide users with detailed guidance services, reduce the learning difficulty, and emphasize the simplicity of BST to make users believe that using BST is easy. Third, HM has a positive effect on attitudes toward the use of BST. If individuals experience a higher sense of wellbeing and pleasure when using a new technology, their attitude towards using it becomes more positive15,16. This implies that if users perceive using BST to be pleasurable, their attitudes towards using it will be more positive. Therefore, enterprises can provide users with interesting and meaningful experiences locally to motivate them to willingly adopt and continue using BST due to the enjoyment they derive from it. Fourthly, FCs have a positive impact on attitudes towards using BST. They can foster users’ positive attitudes and opinions towards new technologies42. This finding suggests that FCs exhibit a positive influence on consumers’ attitudes towards BST when integrated with the UTAUT2 and TPB theoretical frameworks. This indicates that users’ perceptions of BST are shaped by external variables, social conditions, or other fostering circumstances, which significantly contribute to the formation of attitudes towards BST’s utilization and further augment users’ positive appraisals of the technology. Consequently, this enables companies to devise more precise marketing strategies and incentives, as well as optimize and leverage promotional conditions to escalate consumers’ favorable attitude towards BST, ultimately inspiring more users to adopt it. Fifth, PR has a negative effect on attitudes toward the use of BST and discourages consumers from using it. Previous research has shown that knowledge is correlated with perceived risk, and a lack of knowledge about BST leads to perceptual errors regarding their use4. These errors can result in a higher perception of risk and discourage consumers from using or purchasing BST. Therefore, companies should provide easily understandable information about the working principle, advantages, and safety of BST. They should also organize activities and disseminate content about BST through the Internet to expand its influence. This will enhance consumers’ understanding of BST and reduce their perceived potential risks. However, we find that two factors of UTAUT2 (PV and HBs) are not significantly related to attitudes. The first reason for this finding is that the effect of PV on attitudes toward the use of BST is not significant. In a study conducted by Vafaei-Zadeh et al.54, it was also found that PV had a less important role in forming attitudes. There is no linear correlation between PV and ATT, and there may be some moderating factors between price awareness and consumer attitudes55. However, most of the respondents in this survey were young people aged 18 to 30 who are more concerned with other performance factors such as technology54. Therefore, PV does not influence users’ attitudes towards BST use. Secondly, HB is also not a significant predictor of attitude towards BST use. In a previous study16, the results showed that HBs do not have a significant impact on attitudes. This indicates that users’ driving habits are not affected by either conventional fuel vehicles or rechargeable electric vehicles, and their attitude towards using BST is not influenced by their driving habits. Therefore, habit cannot be considered to be one of the factors affecting attitudes towards the use of BST. Theoretical and practical implications This study initially delved into the theories associated with the willingness to accept a new technology and scrutinizes the factors that might sway users’ propensity to adopt BST. To date, research on BST has predominantly revolved around technology implementation8–10, with a relatively scant exploration of the factors influencing individuals’ inclination to employ this technology4,7. Therefore, building upon relevant theories and prior research, we developed a model of user intention to adopt battery swapping based on the Theory of Planned Behavior (TPB) and the User Acceptance Theory of Adoption of New Technologies (UTAUT2), incorporating perceived risk factors. Through the results, we show that five factors, PE, EE, HM, FCs, and PR, have a significant effect on attitudes toward using BST, and three factors, ATT, SBN, and PBC, affect users’ intentions to use BST. In conclusion, this research extends the exploration of factors influencing users’ adoption and utilization of BST and develops a more comprehensive model to assess users’ intentions to employ BST, grounded in the technological acceptance theory framework. This research facilitates a more profound comprehension of users’ perspectives of BST and their intention to utilize the technology within the electric vehicle industry. In the meantime, we can discern some of the factors that contribute to users’ positive appraisal of this technology. In summary, this study broadens the exploration of factors influencing users’ acceptance and adoption of BST, provides insights into the model of perceived risk factors, and builds upon the theoretical framework of Technology Acceptance Model to construct a more comprehensive model for examining users’ intention to use BST. Limitations and future research While this study offers a more comprehensive understanding of users’ intentions to employ BST, certain limitations should be acknowledged. Firstly, the theoretical framework of this study exhibits certain shortcomings. The integration of multiple theories could contribute to the complexity of this research. Thus, to ensure the model’s increased validity and generalizability, it is necessary to re-test it on a larger scale with an enlarged sample size. Secondly, the present study mostly focused on potential users, with little consideration of the current users16. Therefore, future research could focus on studying users who are already using BST. Additionally, our research methodology employed convenience sampling for a questionnaire collection, which resulted in a sample that was not sufficiently representative. Future research could utilize methods such as stratified sampling to enhance the representativeness of the sample. Lastly, based on the investigation of users’ acceptance of BST technology, future research should consider how to provide new experiences to BST users so that they can derive greater pleasure from using BST34. Appendix A AttitudeATT1: I think it is an important thing to use EVs which support battery swap technology. 8,14,30 ATT2: I am interested in EVs which support battery swap technology.ATT3: I think it would be a good idea to use battery swap technology.ATT4: I think it is a good thing to accelerate the purchase of EVs which support battery swap technology.SubjectiveNormSBN1: Family and relatives have bought and used battery swap technology, and maybe I will buy and use it too.SBN2: The advertisements in various media sources about battery swap technology can promote my purchase and use of it.SBN3: If the people around me use battery swap technology, I think I will use it. 8,14,30 PerceivedBehavioralControlPBC1: I have the freedom to decide whether or not to use battery swap technology.PBC2: I will be able to afford to buy an EV which supports battery swap technology in the future.PBC3: If I wanted to, I could definitely choose battery swap technology for my next car purchase. 8,14,30 IntentionToUseITU1: I would consider buying and using an electric vehicle which supports battery swap technology.ITU2: I plan to try to buy and use an electric vehicle which supports battery swap technology.ITU3: I would buy an electric vehicle which supports battery swap technology and recommend it to my colleagues, friends, and family.ITU4: I am very much looking forward to the launch of more different brands of electric vehicles which support battery swap technology.ITU5: I imagine that in the future I will buy and use an EV which supports battery swap technology. 8,14,30 PerformanceExpectancyPE1: Battery swap technology helps me with my work.PE2: Battery swap technology might be kinder to the environment.PE3: Battery swap technology saves me money.PE4: Battery swap technology can improve my work efficiency. 14 EffortExpectancyEE1: Is it easy for me to learn how to use an electric vehicle which supports battery swap technology.EE2: I understand and know how to use battery swap technology.EE3: I think battery swap technology will be easy to use.EE4: I think I can use battery swap technology proficiently. 14 HedonicMotivationHM1: I think it would be interesting to have an electric car which supports battery swap technology.HM2: It seems to me that I would be more comfortable with an electric vehicle which supports battery swap technology.HM3: If battery swap technology can make me comfortable in my life, I will buy it even if it is very expensive.HM4: If I was one of the early adopters of battery swap technology, I would be proud. 14 PriceValuePV1: I think battery swap technology is worth the money.PV2: I think the price of the current battery swap technology for electric vehicles is still quite affordable and reasonable.PV3: With the current electric vehicles which support battery swap technology, it is normal that the price is relatively expensive. 14 HabitHB1: I have gotten used to using conventional fuel cars.HB2: I seem to have a hard time switching to an electric vehicle which supports battery swap technology.HB3: My habit of using a conventional gasoline-powered or battery-charged car has made it impossible for me to switch to battery swap technology.HB4: I could not possibly use an electric vehicle which supports battery swap technology because I feel it is no different from a conventional car. 14 FacilitatingConditionFC1: I know the government is making a big effort to build battery swap stations.FC2: I know that the government is actively offering incentives to purchase battery swap technology.FC3: Advances in technology have made me feel safe using battery swap technology.FC4: In terms of how to use a car with battery swap technology, I think it is not different from other vehicles.PerceivedRiskPR1: I am concerned about the time it will take to find a suitable replacement station and having to wait in line for a replacement.PR2: I am concerned about incurring the depreciation costs of batteries and the need to replace them with new ones, resulting in added extra monetary costs.PR3: I am worried that the device will not provide enough battery capacity.PR4: I am worried about social problems like battery swap disputes. ATT2: I am interested in EVs which support battery swap technology. ATT3: I think it would be a good idea to use battery swap technology. ATT4: I think it is a good thing to accelerate the purchase of EVs which support battery swap technology. Subjective Norm SBN1: Family and relatives have bought and used battery swap technology, and maybe I will buy and use it too. SBN2: The advertisements in various media sources about battery swap technology can promote my purchase and use of it. SBN3: If the people around me use battery swap technology, I think I will use it. Perceived Behavioral Control PBC1: I have the freedom to decide whether or not to use battery swap technology. PBC2: I will be able to afford to buy an EV which supports battery swap technology in the future. PBC3: If I wanted to, I could definitely choose battery swap technology for my next car purchase. Intention To Use ITU1: I would consider buying and using an electric vehicle which supports battery swap technology. ITU2: I plan to try to buy and use an electric vehicle which supports battery swap technology. ITU3: I would buy an electric vehicle which supports battery swap technology and recommend it to my colleagues, friends, and family. ITU4: I am very much looking forward to the launch of more different brands of electric vehicles which support battery swap technology. ITU5: I imagine that in the future I will buy and use an EV which supports battery swap technology. Performance Expectancy PE1: Battery swap technology helps me with my work. PE2: Battery swap technology might be kinder to the environment. PE3: Battery swap technology saves me money. PE4: Battery swap technology can improve my work efficiency. Effort Expectancy EE1: Is it easy for me to learn how to use an electric vehicle which supports battery swap technology. EE2: I understand and know how to use battery swap technology. EE3: I think battery swap technology will be easy to use. EE4: I think I can use battery swap technology proficiently. Hedonic Motivation HM1: I think it would be interesting to have an electric car which supports battery swap technology. HM2: It seems to me that I would be more comfortable with an electric vehicle which supports battery swap technology. HM3: If battery swap technology can make me comfortable in my life, I will buy it even if it is very expensive. HM4: If I was one of the early adopters of battery swap technology, I would be proud. Price Value PV1: I think battery swap technology is worth the money. PV2: I think the price of the current battery swap technology for electric vehicles is still quite affordable and reasonable. PV3: With the current electric vehicles which support battery swap technology, it is normal that the price is relatively expensive. HB1: I have gotten used to using conventional fuel cars. HB2: I seem to have a hard time switching to an electric vehicle which supports battery swap technology. HB3: My habit of using a conventional gasoline-powered or battery-charged car has made it impossible for me to switch to battery swap technology. HB4: I could not possibly use an electric vehicle which supports battery swap technology because I feel it is no different from a conventional car. Facilitating Condition FC1: I know the government is making a big effort to build battery swap stations. FC2: I know that the government is actively offering incentives to purchase battery swap technology. FC3: Advances in technology have made me feel safe using battery swap technology. FC4: In terms of how to use a car with battery swap technology, I think it is not different from other vehicles. Perceived Risk PR1: I am concerned about the time it will take to find a suitable replacement station and having to wait in line for a replacement. PR2: I am concerned about incurring the depreciation costs of batteries and the need to replace them with new ones, resulting in added extra monetary costs. PR3: I am worried that the device will not provide enough battery capacity. PR4: I am worried about social problems like battery swap disputes. Acknowledgements The authors would like to thank all those who participated in this study. Author contributions Xing Xu: Writing–original draft, Writing–review & editing, Methodology, Formal analysis, Funding acquisition.Siyu Feng: Writing–original draft, Writing–review & editing, Visualization, Conceptualization, Software, Data curation.Zhiling He: Writing– review & editing, Methodology, Visualization.Tiansheng Xia: Supervision, Project administration, Review. Funding This study was supported by the Guangdong Province Philosophy and Social Science 2023 Project (GD23XYS016) and the Ministry of Education’s Humanities and Social Sciences Research Program (23YJC760131). Data availability The data supporting the findings of this study are not publicly available due to privacy and ethical restrictions. Anonymized data may be made available from the corresponding author upon reasonable request and subject to institutional and ethical approval. Declarations Competing interests The authors declare no competing interests. Ethics statement This work has been approved by the Departmental Ethics Committee and the Institutional Review Board of the Guangdong University of Technology (No. GDUTXS2023237). Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The data supporting the findings of this study are not publicly available due to privacy and ethical restrictions. Anonymized data may be made available from the corresponding author upon reasonable request and subject to institutional and ethical approval.