Transportation Research Part A 174 (2023) 103742

Contents lists available at ScienceDirect

Transportation Research Part A

journal homepage: www.elsevier.com/locate/tra

Consumer purchase intention of new energy vehicles with an

extended technology acceptance model: The role of
attitudinal ambivalence
Lei Zhang a, Hangyan Tong a, Yuqing Liang b, Quande Qin a, c, *
a
College of Management, Shenzhen University, Shenzhen, 518061, China
b
School of Government, Shenzhen Univeristy, Shenzhen, 518061, China
c
Center for Energy & Environmental Policy Research, Beijing Institute of Technology, Beijing, 100081, China

A R T I C L E I N F O A B S T R A C T

Keywords: The mass diffusion of new energy vehicles (NEVs) in the automobile market depends largely on
Purchase intention consumers’ acceptance behavior. In China, traditional culture induces attitudinal ambivalence in
New energy vehicles consumers’ purchase intention of NEVs. This study investigated the role of attitudinal ambiva­
Attitudinal ambivalence
lence on consumers’ purchase intentions. An extended technology acceptance model (TAM) that
Technology acceptance model
incorporates attitudinal ambivalence is proposed. We conduct two sub-studies to verify the
reliability of the proposed model by measuring subjective and objective attitudinal ambivalence
respectively. The empirical results demonstrated that perceived usefulness, perceived ease of use,
and perceived risk have a significant impact on the purchase intention of NEVs through the partial
mediation of attitudinal ambivalence. Perceived risk and perceived ease of use have the strongest
stimulating effect on attitudinal ambivalence and purchase intention respectively. Finally, we
provide several managerial and policy implications for the advancement of NEVs.

1. Introduction

In China the ever-increasing number of vehicles is causing severe environmental pollution. New energy vehicles (NEVs) are highly
valued by the Chinese government, as they represent a breakthrough in alleviating the energy crisis and reducing air pollution (Zhang
& Qin, 2018). Compared with fossil-fueled vehicles, NEVs can significantly reduce pollutant emissions (Su et al., 2021). The available
types of NEVs can be classified into pure electric vehicles, plug-in hybrid electric vehicles and fuel cell vehicles. From 2015 to 2021,
sales of NEVs in China achieved a substantial growth and ranked first in the world. By the end of 2021, China had sold more than 8
million NEVs, accounting for more than 50% of the global total.1
The early development of China’s NEV industry was driven by the policy support, but policy support is gradually being reduced.
The new development requires the NEV industry to change from slow learning to fast development, and from being policy-driven to
market-driven (Zhang & Qin, 2018). In this process, the stimulation of the private market is key. Therefore, it is very important to
identify consumers’ demands and their concerns, as well as to understand the internal mechanisms that affect the purchase intention of
NEVs.

* Corresponding author.
E-mail address: qinquande@gmail.com (Q. Qin).
1
Data was obtained from “https://www.ce.cn/cysc/newmain/yc/jsxw/202208/28/t20220828_38063584.shtml”.

https://doi.org/10.1016/j.tra.2023.103742
Received 12 February 2020; Received in revised form 25 May 2023; Accepted 10 June 2023
Available online 22 June 2023
0965-8564/© 2023 Elsevier Ltd. All rights reserved.
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

In recent years, research on consumer intentions to purchase NEVs focused on product factors, policy factors, and consumer factors.
A number of studies have identified factors, such as the limited driving cruise range of NEVs, insufficient charging infrastructure, long
charging time, insufficient safety, and uncertain performance as the most common reasons why consumers choose not to buy NEVs
(Carley et al., 2013; Lim et al., 2015; Junquera et al., 2016). In contrast, NEVs also have the advantages of low noise, having fast
acceleration and low center of gravity, and requiring low energy consumption. The preferential policies and subsidies for NEVs
promoted the purchase of NEVs in the past (Chen et al., 2019). Additionally, consumers’ value perception influences their purchase
intention of NEVs (Han et al., 2017). Therefore, consumers’ attitude towards NEVs is often not extreme “like” or “dislike”, but mixed
and uncertain, making the purchase decision falls into a dilemma.
Chinese culture, historically shaped by Confucianism, also has its own specific values. Values that favor the purchase of NEVs
include heaven and man (Tian Ren He Yi) and frugality. In terms of the ecological values, Chinese consumers believe that man and
nature are assumed to be one. The use of NEVs is conducive to the protection of the environment, which is in line with the concept of
the “unity of heaven and man”. Frugality is listed as the top value of traditional Chinese culture (Chinese cultural connection, 1987). In
China, a frugal and simple lifestyle is commonly advocated, as it conforms to the requirement of etiquette (Li) in a Confucian society
and helps emphasize one’s virtue (Zhang, 2020). Considering the Chinese government’s subsidies for the private purchase of NEVs and
the relatively cheap electricity price, many consumers believe that buying a NEV is a beneficial behavior to reduce the costs associated
with purchase and use.
Values inhibiting the purchase of NEVs include conservatism and conservative values (mianzi or ‘face’). Chinese consumers are
generally conservative in their consumption psychology, and are reluctant to accept emerging products or enter new markets (Wang &
Lin, 2009). As NEVs are innovative products, consumers may feel uneasy about technological change and the associated uncertainty,
and thus, hesitate to accept innovation (Edison & Geissler, 2003). In the perception of many Chinese consumers, personal mianzi (or
‘face’), which represents a reputation acquired in life through success and ostentation (Hu, 1944), is also very important. Under the
context of Confucian collectivism, Chinese people commonly hold a kind of interdependent self-concept (Wong & Ahuvia, 1998).
Accordingly, Chinese consumers are inclined to consume visibly to showcase their social status, such as choosing expensive fossil-
fueled vehicles with well-known brands (Wang et al., 2016). The afore-mentioned considerations and consumption concepts cause
consumers to be more cautious when buying NEVs.
Thus, it can be seen that consumers may have both positive and negative affects toward NEVs (Wan et al., 2017), indicating
attitudinal ambivalence. As a psychological phenomenon, attitudinal ambivalence is ubiquitous in ecological consumption, and im­
pacts ecological consumption behavior (Conner et al., 2002; Olsen et al., 2005). This study investigates consumers’ purchase intention
of NEVs from the perspective of attitudinal ambivalence.
One of the characteristics of an emerging product is the application of new technology. The case of NEVs is also that of an emerging
product that applies new technology, such as battery technology and the driving system. The technology acceptance model (TAM) and
its extended model have been designed to study users’ acceptance of information technology (Kamal et al., 2020). This model can
explain the factors that affect consumers’ willingness to accept NEVs from a technological perspective (Wang et al., 2018). In view of
the Chinese cultural background and the unique attributes of NEVs, this study extends the TAM to investigate the internal mechanism,
which influences the purchase intention of NEVs. This extended TAM is also used to test the mediating role of attitudinal ambivalence.
This study makes two contributions to the existing literature. Firstly, consumer psychology is investigated from the perspective of
consumer attitudinal ambivalence, which provides theoretical support for promoting the consumption of NEVs. Secondly, attitudinal
ambivalence is integrated into the TAM, thus forming a new extended model, which is more suitable to study consumers’ purchase
intention of NEVs in the Chinese context.
This paper is organized as follows: Section 2 presents the literature review. Section 3 describes the research model and hypotheses.
Data analysis and results are presented in Section 4. Section 5 and Section 6 focus on the discussion and practical implications. Section
7 presents the conclusion.

2. Literature review

2.1. Purchase intention of new energy vehicles

In modeling of purchase intention of NEVs, many studies have been based on the TAM. Kamarudin et al. (2016) adopted the TAM to
predict the acceptance of electric vehicles based on driver intentions. Wu et al. (2019) used the TAM to study the perceived usefulness,
the perceived ease of use, and environmental concern leading to autonomous adoption intention of electric vehicles. Park et al. (2018)
explored the cognition of drivers of electric vehicles based on the TAM and the characteristics of electric vehicles. An extended TAM
was also used to analyze the impact of consumer cognition, perceived risk, perceived usefulness, and financial incentive policies on
consumers’ intentions to adopt NEVs (Wang et al., 2018). Furthermore, the degree to which commercial drivers accepted the
replacement of fossil-fueled vehicles by light-duty electric vehicles was also investigated (Wolff & Madlener, 2019). The TAM has also
been combined with other theories. For instance, based on both the TAM and the planned behavior theory, Wang and Wang (2013)
combined the characteristics of the purchase behavior of NEVs to build a theoretical model. They used this model in an empirical study
on the factors influencing Beijing residents’ intentions to purchase a NEV.

2.2. Attitudinal ambivalence

Attitudinal ambivalence is defined as the state when a person simultaneously has positive and negative feelings toward an object

2
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

(Thompson et al., 1995; Rothman et al., 2017; Akhtar et al., 2020). A person will produce attitudinal ambivalence if highly positive
affect and highly negative affect coexist, which makes them feel conflict and hard to make decisions (Priester & Petty, 1996). Previous
research has suggested some individual traits associated with greater psychological ambivalence, such as cognitive representations of
the self and emotion, dialectical thinking, personal fear of invalidity and low need for cognition (Thompson et al., 1995; Rothman
et al., 2017).
People often have ambivalent attitudes, including when buying green products (Chang, 2011). On the one hand, consumers feel
that green products are energy-saving and protect the environment, so they have a more positive attitude towards using them (Dean
et al., 2012). On the other hand, consumers may have a negative attitude towards green products because they know little about green
products or they have skepticism toward environmental claims (Mohr et al., 1998).
In general, there are two approaches being adopted to measure it. One is to evaluate the subjective perception of ambivalence
directly by asking individuals whether their attitudes are one-sided or mixed toward an attitude object (Priester & Petty, 1996). This
approach reflects the extent to which a person experiences conflict about his/her ambivalent attitude (Itzchakov & Van Harrevel,
2018). However, this approach has been challenged. By means of asking people whether they feel ambivalent, it can be misunderstood
as a question about feeling indecisive or uncertain; thus, may not be able to access the actual subjective experience of ambivalence
(Russell et al., 2011). Additionally, due to the unconsciousness of conflict, it is difficult for people to explicitly express their ambivalent
views as divided (Hass et al., 1992), leading to poor accuracy of subjective approach. The other approach is to evaluate objective
ambivalence, which refers to the co-existence of positive and negative attributes with regard to an attitude object (Kaplan, 1972;
Itzchakov & Van Harrevel, 2018). The objective approach requires an assessment of one’s positive and negative components of attitude
separately, then combine the two components into an ambivalence index through mathematical models. The most commonly used
mathematical model is Griffin’s formula, which is calculated as follows: A= (P + N)/2-|P-N|+C, where A represents attitudinal
ambivalence, P represents positive affect, N represents negative affect and C is a constant to ensure that the value of attitudinal
ambivalence is not negative (Thompson et al., 1995). In this formula, both P and N are non-negative values. Though sometimes
negative affect is measured by non-positive value, N is assigned the absolute value of the negative affect score when calculating A
(Pang et al., 2017). The Griffin’s formula effectively captures the two key components of attitudinal ambivalence as defined by
Thompson et al. (1995). Specifically, |P - N| represents the similarity between positive and negative affect, while (P + N)/2 indicates
the intensity of positive and negative affect. In instances of high attitudinal ambivalence, positive and negative affect are similar in
magnitude, resulting in a small value for |P - N|. Additionally, both positive and negative affect are intense, leading to a large value for
(P + N)/2. Consequently, the value of A is large, indicating a high degree of attitudinal ambivalence. Conversely, when an individual
holds a clear attitude toward an object, there is no similarity between positive and negative affect, resulting in a large value for |P - N|.
Furthermore, the intensity of positive affect is high, while the intensity of negative affect is low (or vice versa), resulting in a value that
is not large for (P + N)/2. Therefore, the value of A is small, indicating a low degree of attitudinal ambivalence. In cases where an
individual has no feelings toward an object, both positive and negative affect have similar magnitudes, but their intensities are low.
This leads to small values for both |P - N| and (P + N)/2, resulting in a small value for A. In summary, the higher the value of A, the
higher the degree of attitudinal ambivalence. Objective approach covers the various components of attitudinal ambivalence more
comprehensively and helps improve the accuracy of inferring the experience of ambivalence (Jonas et al., 1997).
Attitudinal ambivalence is a confused and uncertain state, which can stimulate a consumer’s need for cognitive closure (desire for a
confirmed but not exactly a right decision). For NEVs, a kind of new and high-tech product with the characteristic of public benefit,
consumers will usually have attitudinal ambivalence and thus inhibit their intention of purchasing. In order to compare the predictive
validity of the two measures of ambivalence in the context of purchase of NEVs, both subjective and objective approaches used in this
study.

3. Research model and hypotheses

3.1. The technology acceptance model

The TAM (Davis, 1989) is a theory to both predict and explain the degree to which users accept technology. The TAM includes two
main decisive factors: Perceived usefulness (PU) and perceived ease of use (PEU). According to this model, a consumer’s adoption of a
certain system or a certain technology relies on behavioral intention, which is affected by perceived usefulness and perceived ease of
use.
As an emerging product with new technologies, NEVs adopt battery technology, a novel automobile drive system, and other new
technologies. It is appropriate to adopt TAM to conduct in-depth research on the purchase intentions of NEVs. Considering consumers’
concerns about NEVs, this study integrates perceived risk theory with the TAM, using perceived usefulness, perceived ease of use, and
perceived risk as the common antecedents.
Regarding fossil-fueled vehicles, consumer behavior has been investigated from the perspectives of consumer preference (Liu &
Shankar, 2015), brand image (Apak et al., 2012), brand relationship (Odekerken-Schröder et al., 2010) and special purchase channels
(Kulkarni et al., 2012). However, the NEV market development and technology levels remain immature. Mass market consumers also
have no relevant experience and knowledge (Gärling & Thøgersen, 2001). Consumers are more likely to face attitudinal ambivalence
about whether to buy NEVs. The speed of promoting NEVs’ sale relies largely on the attitudes of consumers (Schuitema et al., 2013),
and attitudinal ambivalence significantly impact consumers’ purchase intention (Hänze, 2001).

3
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

3.2. The effect of perceived usefulness, perceived ease of use and perceived risk on purchase intention

Perceived usefulness represents the degree of profit gained from a certain behavior. This includes both tangible and intangible
material benefits, such as spirit, mentality, and social status (Davis, 1989). NEVs are regarded as a promising technology in the field of
road transportation for the next half century. NEVs can reduce the direct demand for oil, and will play a positive role in environmental
protection (Ou et al., 2010). Several scholars have suggested that NEVs offer clear economic benefits, such as reducing or eliminating
gasoline use and thus, reducing fuel costs (Chen et al., 2019; Kihm & Trommer, 2014). Additionally, the Chinese government has
adopted a series of favorable policies for NEVs. These policies include subsidies, free license plate registration in cities, free parking,
access to restricted areas in cities, and access to high-occupancy vehicle lanes. These policies have played an important role in the
promotion of NEV sales (Wang et al., 2018), and have helped to improve travel efficiency and quality of life for the owners of NEVs.
Following the scale used by Wang et al. (2018), in the present study, perceived usefulness incorporates three aspects: protecting the
environment, reducing household transportation costs, and improving travel efficiency and quality of life.
Perceived ease of use refers to the degree to which an individual subjectively considers it effortless to perform a particular behavior
(Davis, 1989). For NEVs, accessing to license plates easily is one of the perceived ease of use. For example, in some Chinese cities with
severe traffic-restriction, NEV owners can obtain a special license plate without waiting for a lottery. Referring to Wu et al. (2019) and
considering the technical and performance aspects of NEVs, this study investigates the factors for purchasing NEVs as follows: ease of
applying for a license plate, ease of travel, and ease of operation.
In this study, two internal beliefs, perceived usefulness and perceived ease of use, are considered as antecedents of support for the
purchase intention of NEVs (Globisch et al., 2018). The following two original hypotheses are proposed:
H1: Perceived usefulness has a positive impact on purchase intention.
H2: Perceived ease of use has a positive impact on purchase intention.
The traditional approach of TAM only considers the positive factors of new technologies and neglects the negative perceptions of
users (Wang et al., 2020). Perceived risk is the main barrier that directly or indirectly affects the purchase intention of NEVs (Jaiswal
et al., 2021).
Perceived risk refers to the uncertainty and negative outcomes that consumers experience when they buy goods (Dunn et al., 1986).
Perceived risk varies across products (Featherman & Pavlou 2003). Wang et al. (2020) pointed out that the perceived risks of buying
and owning NEVs can be divided into six categories: financial risk, physical risk, time risk, performance risk, social risk, and con­
venience risk.
In China, the financial risk associated with NEVs includes their comparatively high maintenance costs and low residual used value
(Brand et al., 2017). NEVs are emerging products; therefore, several technologies are still underdeveloped compared with fossil-fueled
vehicles, such as their limited driving range (Wang et al., 2017a), and long charging time (Schuitema et al., 2013). As a result,
consumers consider the risk of convenience before buying NEVs, including time loss, and the disappointed functions of the products
(Featherman & Pavlou, 2003). Of course, the convenience of having a charging facility at home is important to the potential buyers of
NEVs (Qian et al., 2019). Insufficient charging infrastructure is inconvenient for consumers (Wang et al., 2017b). Consumer safety
assurance (e.g., spontaneous combustion) is a further problem for NEVs (Brand et al., 2017; Flamm & Agrawal, 2012). Mianzi
awareness is a dimension of Confucian values that may influence the decision of Chinese consumers to adopt NEVs, especially
regarding their symbolic significance (Qian & Yin, 2017). In Chinese society, most consumers are still more willing to buy fossil-fueled
vehicles, as in this way, they can show their social identity, social status, and self-image (Wang et al., 2016). Consumers who buy NEVs
might worry that the use of NEVs may be misunderstood by their relatives and friends (Wang et al., 2018). The following hypotheses
are proposed:
H3: Perceived risk has a negative impact on purchase intention.

3.3. The effect of perceived usefulness, perceived ease of use and perceived risk on attitudinal ambivalence

Previous studies have demonstrated that an individual’s perceived usefulness and perceived ease of use affect attitude towards the
object (Wu & Chen, 2017). Chang (2011) found that perceived quality is a significant predictor of attitudinal ambivalence toward
green products, while perceived utility is a salient predictor. Products with high perceived ease of use are often regarded as trustworthy
(Vahdat et al., 2021). The following two hypotheses are proposed.
H4: Perceived usefulness has a negative impact on attitudinal ambivalence.
H5: Perceived ease of use has a negative impact on attitudinal ambivalence.
Perceived risk reflects the unpredictability of the buying process and is often seen as a negative insight (Laroche et al., 2005), which
is positively associated with expression ambivalence when consumers want to buy green goods (Le et al., 2019). This study hy­
pothesizes that:
H6: Perceived risk has a positive impact on attitudinal ambivalence.

3.4. Attitudinal ambivalence on purchase intention

Wang and Wu (2015) suggested that green purchase affect presents significant two-dimensional characteristics, namely positive
and negative affect-based dimensions, which impose different influences on green purchase behavior. Positive affect can promote
green purchasing behavior intention, while negative affect can disturb and diminish it (Chan & Lau, 2000).
Attitudinal ambivalence may either directly or indirectly influence consumers’ behaviors both before and during a purchase (Otnes

4
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

et al., 1997). On the one hand, consumers have strong positive affect about NEVs; they believe that NEVs are beneficial for envi­
ronmental protection, will receive government subsidies, are easily licensed and have no driving restrictions. On the other hand,
consumers also have strong negative affect; they may worry that NEVs are not convenient to charge, may have the poor driving ability,
and may be unsafe. Attitudinal ambivalence has a negative impact on individual behavioral intentions (Hu et al., 2017), which also
reduces consumers’ brand satisfaction and brand loyalty (Olsen et al., 2005). The following hypothesis is thus proposed:
H7: Attitudinal ambivalence has a negative impact on the purchase intention of NEVs.
These hypotheses were tested through two sub-studies. In sub-study 1 and sub-study 2, attitudinal ambivalence was measured by
objective and subjective approach, respectively.
Fig. 1 summarizes the research model of the purchase intention of NEVs as proposed by this study.

4. Data analysis and results

4.1. Data analysis and results of sub-study 1

4.1.1. Sample and data collection

Data for sub-study 1 were collected in two ways between June 8th to September 23th in 2019. One was by distributing the
questionnaires on-site at automobile trading markets and vehicle administration offices, and the other was by distributing the ques­
tionnaires online via Credamo, a paid survey platform specializing in scientific research. The on-site survey was conducted in
Guangdong province and the online survey covers 20 provinces in China. A total of 602 questionnaires were recovered, among which
163 on-site and 419 online were valid. SPSS 22.0 and AMOS 22.0 software packages were used to analyze the data and confirmatory
factor analyses were conducted to evaluate the samples. Table 1 presents the sample distribution.

4.1.2. Variables and measurement

The questionnaire consisted of two parts. The first part collected basic demographic information about the respondents, and the
second part collected data on the potential constructs. In this sub-study, all measured items were based on the literature review.
Through Chinese-English translation and semi-structured, small-scale personal interviews, the variable content and text expressions
were modified to adapt to the current research environment. Two marketing experts were invited to evaluate the content of the scale.
Finally, a 5-point Likert scale (where 1 = strongly disagree and 5 = strongly agree) was used to measure six variables. The measured
constructs are shown in Table 2 and the items were asked in the order shown in the table. Specifically, objective attitudinal ambiv­
alence was measured by calculating positive affect and negative affect scores using Griffin’s formula (Thompson et al., 1995), and C
was assigned a value of 0. In this and the following tables, “Mean” indicates the mean of the score of each indicator, and “SD” stands for
the standard deviation.

4.1.3. Measurement model analysis

The Cronbach’s alpha coefficient, the corrected overall correlation coefficient (CITC), and the combined reliability value (CR) were
used to test the reliability of the scale. The rotated component matrix loading and the average variance extracted (AVE) were used to
test the validity of the scale (as shown in Table 3). Both rotated component matrix loading and CITC exceeded 0.5, Cronbach’s alpha
exceeded 0.7, CR exceeded 0.7, and AVE exceeded 0.5. Furthermore, as shown in Table 4, the square root of AVE (i.e., on the diagonal)
in bold for each construct exceeded its correlations with other constructs, indicating acceptable discriminant validity for the con­
structs. This indicates that the scale passed the reliability and validity tests, and the questionnaire has both good reliability and
validity.

4.1.4. Structural model analysis

The framework presented in this sub-study was analyzed and verified via the structural equation model (SEM; see Fig. 1).
Considering the possible influence of the common method bias (CMB) on the results of this sub-study, the Harman’s one-factor test was
conducted to evaluate the common method bias. The results show that it passed the test.

Fig. 1. The proposed research model of the purchase intention of NEVs.

5
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

Table 1
Demographic profile of respondents of sub-study 1.
Demographic variable Frequency Percentage (%)

Gender Male 298 51.2

Female 284 48.8
Age 18–24 163 28.0
25–34 245 42.1
35–44 150 25.8
Above 45 24 4.1
Educational level Junior high or below 2 0.3
High school or technical secondary school 7 1.2
Associate degree 81 13.9
Bachelor’s degree 406 69.8
Master’s degree or above 86 14.8
Job Employees of enterprises and public institutions or managers 200 34.4
Technicians 124 21.3
Civil servants 35 6.0
Individual proprietor 102 17.5
Student 108 18.6
Other 13 2.2
Family size Two persons or below 92 15.8
Three persons 278 47.8
Four persons 156 26.8
Five persons or above 56 9.6
Total annual Below 100,000 RMB 41 7.0
household income 100,000 ~ 200,000 RMB 292 50.2
200,000 ~ 500,000 RMB 148 25.4
Above 500,000 RMB 101 17.4
Does the family currently own a car? Yes 426 73.2
No 156 26.8
Does the family plan to buy a car in the next five years? Yes, I plan to buy a fossil-fueled vehicle 176 30.2
Yes, I will consider buying a NEV 320 55.0
No 86 14.8
Total 582 100

Table 2
Measured constructs of sub-study 1.
Construct Indicator Measurement items Literature Mean SD

Perceived PU1 NEVs are useful to decrease carbon emissions and alleviate energy shortage Wang et al., 2018 3.96 0.977
usefulness problems.
(PU) PU2 NEVs are useful to decrease my household expenditures on transportation. 4.01 0.899
PU3 NEVs can improve my travel efficiency and my living quality. 4.00 1.060
Perceived ease of PEU1 I think it is easy to obtain a license plate for NEVs. Wu et al., 2019 3.92 1.134
use (PEU) PEU2 I think it would be easy for me to drive NEVs to anywhere I want. 3.95 0.999
PEU3 I think it will be easy to use NEVs. 3.84 0.966
Perceived risk PR1 I am afraid of suffering financial losses when using NEVs (because of high Li et al., 2017;Wang 3.10 1.310
(PR) maintenance cost and second-hand car salvage value). et al., 2020
PR2 I would not feel very safe when I drive a NEV on the road. 2.98 1.290
PR3 Considering the disadvantages of NEVs (e.g., limited driving range, long time of 3.10 1.312
recharge, and insufficient charging infrastructure), I think that using NEVs may
incur considerable time losses.
PR4 I worry about whether NEVs will really perform as well as fossil-fueled 2.87 1.212
vehicles.
PR5 I worry that my family and friends may not understand my decision if I 2.84 1.227
purchase a NEV.
PR6 I worry that driving a NEV will not be convenient for me. 2.87 1.223
Positive affect POS1 NEVs are attractive to me. Barrett & Russell, 1999; 3.98 0.861
(POS) POS2 I am passionate about NEVs. Watson et al., 1988 3.84 1.059
POS3 I feel proud to possess a NEV. 3.88 0.963
Negative affect NEG1 I hate NEVs. 1.92 0.867
(NEG) NEG2 I am worried about NEVs. 2.88 1.154
NEG3 I am disappointed by NEVs. 1.97 0.898
Purchase PI1 I would consider buying NEVs. Dodds et al., 1991Wang 3.76 1.036
intention (PI) PI2 I would like to buy a NEV. et al., 2017c 3.81 0.89
PI3 I will recommend NEVs to others. 3.85 0.904

6
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

Table 3
Reliability and validity analysis of variables of sub-study 1.
Constructs Indicator Rotated component CITC Cronbach’s α CR AVE
matrix loading

PU PU1 0.850 0.718 0.849 0.851 0.655

PU2 0.836 0.700
PU3 0.866 0.744
PEU PEU1 0.868 0.769 0.844 0.849 0.653
PEU2 0.833 0.691
PEU3 0.808 0.684
PR PR1 0.801 0.720 0.901 0.902 0.604
PR2 0.827 0.754
PR3 0.815 0.742
PR4 0.800 0.715
PR5 0.818 0.739
PR6 0.799 0.713
PI PI1 0.586 0.721 0.861 0.865 0.682
PI2 0.581 0.723
PI3 0.623 0.778

Table 4
Means, standard deviations, and correlations of constructs of sub-study 1.
Construct Mean SD PU PEU PR PI

PU 3.990 0.860 0.809

PEU 3.904 0.904 0.288*** 0.808
PR 2.962 1.033 − 0.018 − 0.148 0.777
PI 3.806 0.836 0.500*** 0.626*** − 0.448*** 0.826

Note: *** shows significance level of 1%.

Bolded diagonal elements are the square root of AVE in Table 3.

Prior to the analysis, a goodness of fit index was calculated to ensure that the proposed model is suitable for SEM and analysis.
Through data analysis, the following fitting indexes of this model and sample data were set: x2/df = 1.188, RMSEA = 0.018, GFI =
0.977, AGFI = 0.966, NFI = 0.978, IFI = 0.996, CFI = 0.996, PGFI = 0.682, and PNFI = 0.774. The x2/df was lower than 3, RMSEA was
lower than 0.08, GFI, AGFI, NFI, CFI, and IFI exceeded 0.9, and both PGFI and PNFI exceeded 0.5. The squared multiple correlation of
AA and PI are 0.353 and 0.739 respectively, which demonstrated that the model has a good fit and predictive ability. The standardized
path coefficients of this sub-study are shown in Table 5.
Table 6 shows the mediating effect test results of AA. The bootstrap method was used to test the mediating effect, and thus we
generated 5000 resamples. The point estimators in Table 6 are obtained by multiplying the two coefficients of the meditating paths
from PU, PEU and PR through AA to PI as shown in Table 5. In the confidence interval test, whether the magnitude of the effect is
significant is judged by whether the 95% confidence interval contains 0. Table 6 shows that PU has a significant positive mediating
effect on PI, with 95% confidence interval [0.015, 0.077] and [0.015, 0.076] excluding 0. PEU has a significant positive mediating
effect on PI, and 95% confidence intervals [0.102, 0.179] and [0.101, 0.177] excluding 0. PR has a significant negative mediating
effect on PI, with 95% confidence interval [-0.199, − 0.115] and [-0.198, − 0.114] excluding 0. At the same time, the absolute value of
the ratio (Z) of the estimators of each effect point to the standard error SE being greater than 1.96, which further verified the test results
of the Bootstrap method. Therefore, it can be concluded that attitudinal ambivalence plays a partial mediating role in the influence of
perceived usefulness, perceived ease of use, and perceived risk on purchase intention. The final formation of the purchase intention
influence mechanism model of NEVs is shown in Fig. 2.

Table 5
Standardized path coefficient of sub-study 1.
Standard S.E. C.R. P value
estimate

AA <— PU − 0.113 0.053 − 2.848 0.004

AA <— PEU − 0.350 0.067 − 8.456 0.000
AA <— PR 0.391 0.045 10.053 0.000
PI <— AA − 0.393 0.026 − 10.407 0.000
PI <— PU 0.314 0.032 9.017 0.000
PI <— PEU 0.330 0.043 8.540 0.000
PI <— PR − 0.218 0.028 − 6.285 0.000

7
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

Table 6
Mediating effect test of sub-study 1.
Point SE Z Bootstrap test results
estimator
Bias-corrected test Percentile test

Lower confidence Upper confidence Lower confidence Upper confidence

limit limit limit limit

The mediation PU → PI 0.044 0.016 2.75 0.015 0.077 0.015 0.076

effect PEU → 0.138 0.019 7.26 0.102 0.179 0.101 0.177
PI
PR → PI − 0.154 0.021 − 7.33 − 0.199 − 0.115 − 0.198 − 0.114

Fig. 2. Model diagram of sub-study 1.

4.2. Data analysis and results of sub-study 2

4.2.1. Sample and data collection

Data for sub-study 2 were collected through an online survey by Credamo from July 15th to July 29th in 2022, with respondents
sampled from 29 provinces in China. A total of 1023 questionnaires were collected, of which 627 were valid. Table 7 presents the
sample distribution.

4.2.2. Variables and measurement

In this sub-study, the questionnaire is also consisted of two parts. The first part collected basic demographic information about the
respondents, and the second part collected data on the potential constructs. The items were asked in the order shown in Table 8. Most
of the measurement items are the same as sub-study 1, the difference is that attitudinal ambivalence is measured through several
questions based on Olsen et al (2009) and Itzchakov & Van Harreveld (2018).

4.2.3. Measurement model analysis

As shown in Table 9 and Table 10, the scale passed the reliability and validity tests, and the questionnaire has both good reliability
and validity.

4.2.4. Structural model analysis

Through data analysis, the following fitting indexes of this model and sample data were set: x2/df = 2.015, RMSEA = 0.040, GFI =
0.955, AGFI = 0.938, NFI = 0.956, IFI = 0.977, CFI = 0.977, PGFI = 0.698, and PNFI = 0.781. The x2/df was lower than 3, RMSEA was
lower than 0.08, GFI, AGFI, NFI, CFI, and IFI exceeded 0.9, and both PGFI and PNFI exceeded 0.5. The squared multiple correlation of
AA and PI are 0.553 and 0.440 respectively. The standardized path coefficients of this sub-study are shown in Table 11.
Table 12 shows the mediating effect test results of AA. PU has a significant positive mediating effect on PI, with 95% confidence

8
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

Table 7
Demographic profile of respondents of sub-study 2.
Demographic variable Frequency Percentage (%)

Gender Male 289 46.1

Female 337 53.7
Age 18–24 130 20.7
25–34 349 55.7
35–44 93 14.8
Above 45 55 8.8
Educational level Junior high or below 44 7.0
High school or technical secondary school 63 10.0
Associate degree 80 12.8
Bachelor’s degree 388 61.9
Master’s degree or above 52 8.3
Job Employees of enterprises and public institutions or managers 275 43.9
Technicians 124 19.8
Civil servants 41 6.5
Individual proprietor 42 6.7
Student 86 13.7
Other 59 9.4
Family size Two persons or below 44 7.0
Three persons 279 44.5
Four persons 165 26.3
Five persons or above 139 22.2
Total annual Below 100,000 RMB 118 18.8
household income 100,000 ~ 200,000 RMB 299 47.7
200,000 ~ 500,000 RMB 170 27.1
Above 500,000 RMB 40 6.4
Does the family currently own a car? Yes 532 84.8
No 95 15.2
Does the family plan to buy a car in the next five years? Yes, I plan to buy a fossil-fueled vehicle 91 14.5
Yes, I will consider buying a NEV 446 71.1
No 90 14.4
Have the family ever Yes 116 18.5
bought a NEV? No 511 81.5
Total 627 100

Table 8
Measured constructs of sub-study 2.
Construct Indicator Measurement items Literature Mean SD

Perceived PU1 NEVs are useful to decrease carbon emissions and alleviate energy Wang et al., 2018 4.28 0.781
usefulness shortage problems.
(PU) PU2 NEVs are useful to decrease my household expenditures on 4.16 0.856
transportation.
PU3 NEVs can improve my travel efficiency and my living quality. 3.93 0.886
Perceived ease of PEU1 I think it is easy to obtain a license plate for NEVs. Wu et al., 2019 3.79 0.968
use (PEU) PEU2 I think it would be easy for me to drive NEVs to anywhere I want. 3.93 0.989
PEU3 I think it will be easy to use NEVs. 3.94 0.926
Perceived risk (PR) PR1 I am afraid of suffering financial losses when using NEVs (because of high Li et al., 2017;Wang et al., 3.37 1.095
maintenance cost and second-hand car salvage value). 2020
PR2 I would not feel totally safe when I drive a NEV on the road. 3.20 1.104
PR3 Considering the disadvantages of NEVs (e.g., limited driving range, long 3.16 1.125
time of recharge, and insufficient charging infrastructure), I think that
using NEVs may incur considerable time losses.
PR4 I worry about whether NEVs will really perform as well as fossil-fueled 3.16 1.098
vehicles.
PR5 I worry that my family and friends may not understand my decision if I 2.74 1.125
purchase a NEV.
PR6 I worry that driving a NEV will not be convenient for me. 2.85 1.108
Attitudinal AA1 I have mixed feelings about NEVs. Olsen et al., 2009Itzchakov & 2.82 1.040
ambivalence AA2 I feel conflicted regarding my attitude towards NEVs. Van Harreveld, 2018 2.95 1.106
(AA) AA3 I have mixed feelings about buying NEVs. 3.09 1.124
Purchase intention PI1 I would consider buying NEVs. Dodds et al., 1991Wang 3.79 1.015
(PI) PI2 I would like to buy a NEV. et al., 2017c 3.77 1.024
PI3 I will recommend NEVs to others. 3.70 1.043

9
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

Table 9
Reliability and validity analysis of variables of sub-study 2.
Constructs Indicator Rotated component CITC Cronbach’s α CR AVE
matrix loading

PU PU1 0.825 0.696 0.831 0.833 0.625

PU2 0.856 0.715
PU3 0.787 0.665
PEU PEU1 0.834 0.687 0.834 0.835 0.627
PEU2 0.851 0.721
PEU3 0.779 0.676
PR PR1 0.711 0.612 0.860 0.860 0.507
PR2 0.751 0.659
PR3 0.743 0.661
PR4 0.778 0.703
PR5 0.727 0.600
PR6 0.746 0.667
AA AA1 0.753 0.691 0.829 0.830 0.620
AA2 0.755 0.700
AA3 0.781 0.673
PI PI1 0.852 0.799 0.891 0.892 0.734
PI2 0.846 0.804
PI3 0.794 0.755

Table 10
Means, standard deviations, and correlations of constructs of sub-study 2.
Construct Mean SD PU PEU PR AA PI

PU 4.123 0.728 0.791

PEU 3.890 0.833 0.466*** 0.792
PR 3.081 0.850 − 0.230*** − 0.208*** 0.712
AA 2.956 0.942 − 0.494*** − 0.477*** 0.610*** 0.787
PI 3.754 0.930 0.478*** 0.515*** − 0.424*** − 0.563*** 0.857

Note: *** shows significance level of 1%.

Bolded diagonal elements are the square root of AVE in Table 9.

Table 11
Standardized path coefficient of sub-study 2.
Standard S.E. C.R. P value
estimate

AA <— PU − 0.263 0.061 − 5.779 0.000

AA <— PEU − 0.251 0.056 − 5.540 0.000
AA <— PR 0.497 0.049 11.136 0.000
PI <— AA − 0.218 0.063 − 3.386 0.000
PI <— PU 0.196 0.064 4.054 0.000
PI <— PEU 0.280 0.059 5.763 0.000
PI <— PR − 0.187 0.057 − 3.589 0.000

Table 12
Mediating effect test of sub-study 2.
Point SE Z Bootstrap test results
estimator
Bias-corrected test Percentile test

Lower confidence Upper confidence Lower confidence Upper confidence

limit limit limit limit

The mediation PU → PI 0.057 0.021 2.714 0.022 0.105 0.019 0.102

effect PEU → 0.055 0.022 2.500 0.019 0.104 0.017 0.100
PI
PR → PI − 0.108 0.039 − 2.769 − 0.191 − 0.039 − 0.190 − 0.037

interval [0.022, 0.105] and [0.019, 0.102] excluding 0. PEU has a significant positive mediating effect on PI, and 95% confidence
intervals [0.019, 0.104] and [0.017, 0.100] excluding 0. PR has a significant negative mediating effect on PI, with 95% confidence
interval [-0.191, − 0.039] and [-0.190, − 0.037] excluding 0. At the same time, the absolute value of the ratio (Z) of the estimators of

10
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

each effect point to the standard error SE being greater than 1.96. Therefore, it can be concluded that attitudinal ambivalence plays a
partial mediating role in the influence of perceived usefulness, perceived ease of use, and perceived risk on purchase intention. The
final formation of the purchase intention influence mechanism model of NEVs is shown in Fig. 3.

5. Discussion

Perceived usefulness, perceived ease of use and perceived risk have a significant impact on the purchase intention of NEVs, of which
perceived usefulness and perceived ease of use have significant positive effects and perceived risk has a significant negative effect.
Attitudinal ambivalence plays a partial mediating role, the perception of the core performance of innovative products can reduce
attitudinal ambivalence, while the perception of risks of different aspects of innovative products can stimulate attitudinal ambivalence.
It is worth noting that attitudinal ambivalence amplifies the direct effects of perceived usefulness, perceived ease of use, and perceived
risk on the purchase intention of NEVs. That is, the positive indirect effects of perceived usefulness and perceived ease of use essentially
make their positive direct effects more positive, and the negative indirect effect of perceived risk makes its negative direct effect more
negative.
From sub-study 1 and sub-study 2, perceived risk and perceived ease of use have the strongest effect on attitudinal ambivalence and
purchase intention of NEVs, respectively. The former reflects that perceived risk is still an important factor in motivating attitudinal
ambivalence and thus restricting consumers’ purchase intention of NEVs, while the latter implies that consumers gradually understand
and accept the functional attributes of NEVs, and the convenience in using them is conducive to the promotion of NEVs.
Comparing sub-study 1 and sub-study 2, the two sets of findings are qualitatively similar, indicating that both direct subjective
attitude measures of ambivalence and indirect positive/negative affect measures of ambivalence are reliable in the consumption field
of NEVs.

6. Practical implications

This study measured subjective and objective attitudinal ambivalence separately and verified the significant influence of attitudinal
ambivalence on the purchase intention of NEVs. High attitudinal ambivalence can lead to lower purchase intention, thus reducing
consumers’ attitudinal ambivalence contributes to the widespread adoption of NEVs.
Over the past 40 years, the Chinese government has played a pivotal and proactive role in driving economic development. The new
energy industry is a new area which is initiated and supported strongly by the government. The changes of the related industrial
policies push the NEV enterprises to grow rapidly and engage in fierce market competition directly. In view of the different effects of
consumers’ perceived usefulness, perceived ease of use, and perceived risk of NEVs on attitudinal ambivalence and purchase intention,
some measures may be adopted by government to actively address it. For this purpose, some practical implications can be adopted as
follows:
As perceived risk has the greatest impact on attitudinal ambivalence, suffering financial losses can be the primary concern of
consumers. With the increase of the market penetration of NEVs, the after-sales service of new energy vehicles is highly valued. As for
high maintenance cost, the government can encourage more relevant enterprises to join technical research and development in the
fields of expendable auto parts, body materials etc. One important issue is the second-hand NEVs evaluation. Currently, China lacks a
unified evaluation standard for the residual value in the second-hand NEVs market. A second-hand NEV authentication and evaluation

Fig. 3. Model diagram of sub-study 2.

11
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

system facilitates formulating the market regulation of second-hand NEVs transactions, and maintaining the interests and rights of
consumers. Safety is an imperative concern that necessitates the reinforcement of test verification and offline inspection requirements
for NEVs in order to enhance product reliability.
Perceived ease of use exerts the most significant influence on the purchase intention of NEVs. While direct economic incentives may
be diminishing, there is currently a diverse array of non-economic and indirect economic incentive policies in place. These policies
include the dual-credit policy, exemption from license fees, rescission of purchase restrictions (license plate control), and relaxation of
driving restrictions. Properly designing and strategically aligning these policies can enhance consumers’ perception of the convenience
associated with NEVs.

7. Concluding remarks

The sustainable development of China’s NEV industry heavily relies on increased recognition and support from an expanding
consumer base. This study has investigated the factors influencing consumers’ purchase intention of NEVs in China, focusing spe­
cifically on attitudinal ambivalence. The findings highlight that perceived usefulness, perceived ease of use, and perceived risk have
varying degrees of impact on attitudinal ambivalence. Furthermore, attitudinal ambivalence partially mediates the relationship be­
tween consumers’ perception of the advantages and disadvantages of NEVs and their purchase intention. These insights shed light on
the underlying mechanisms that shape consumers’ attitudes and intentions towards NEVs, providing valuable implications for the
industry’s sustainable growth.
The primary limitation of this study is the absence of testing for external validity. In order to validate the research model, future
studies should employ various methods and examine different populations. As the current study focused on purchase intention as the
dependent variable, further investigation into the transition from intention to actual purchase behavior is warranted. Additionally,
exploring alternative measurements of attitudinal ambivalence and conducting longitudinal surveys can help to uncover the causal
relationship between cognitive-affects. These avenues will be explored in future research to enhance the understanding of the topic.
Another limitation of this study is the relatively young age structure of the sample. More than 70% of the respondents in both sub-
study 1 and sub-study 2 are under the age of 35, which implies that the model primarily represents young adults. To enhance the
representativeness of future research, we plan to employ multiple investigation methods.
Furthermore, there is a possibility of automatic response biases, such as Yea-saying effects and Halo effects, due to the non-mixed
order of items associated with the various constructs of the study. In future research, we will address these biases and consider
alternative item sequencing to minimize their influence.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Data availability

Data will be made available on request.

Acknowledgements

The authors thank anonymous referees and an editor of this journal for their valuable comments. This paper is partly supported by
the National Natural Science Foundation of China (Nos. 71871146, 72104146), Guangdong Special Support Program for Young Top-
notch Talent in Science and Technology Innovation (2019TQ05L989), the Planning Project of Philosophy and Social Science in
Shenzhen (No. SZ2022B018) and Shenzhen Science and Technology Program (No. JCYJ20210324093414039) and NTUT-SZU Joint
Research Program (No.2023005).
.

References

Akhtar, N., Siddiqi, U.I., Akhtar, M.N., Usman, M., Ahmad, W., 2020. Modeling attitude ambivalence and behavioral outcomes from hotel reviews. Int. J. Contemp.
Hosp. Manag. 32 (9), 2831–2855.
Apak, S., Göğüş, G.G., Karakadılar, İ.S., 2012. An analytic hierarchy process approach with a novel framework for luxury car selection. Procedia Soc. Behav. Sci. 58,
1301–1308.
Barrett, L.F., Russell, J.A., 1999. The structure of current affect: controversies and emerging consensus. Curr. Dir. Psychol. Sci. 8 (1), 10–14.
Brand, C., Cluzel, C., Anable, J., 2017. Modeling the uptake of plug-in vehicles in a heterogeneous car market using a consumer segmentation approach. Transp. Res. A
Policy Pract. 97, 121–136.
Carley, S., Krause, R.M., Lane, B.W., Graham, J.D., 2013. Intent to purchase a plug-in electric vehicle: a survey of early impressions in large us cites. Transp. Res. Part
D: Transp. Environ. 18, 39–45.
Chan, R.Y.K., Lau, L.B.Y., 2000. Antecedents of green purchases: a survey in China. J. Consum. Mark. 17 (4), 338–357.
Chang, C., 2011. Feeling ambivalent about going green. J. Advert. 40 (4), 19–32.
Chen, K., Ren, C.R., Gu, R., Zhang, P.D., 2019. Exploring purchase intentions of new energy vehicles: from the perspective of frugality and the concept of “mianzi”.
J. Clean. Prod. 230, 700–708.
Chinese cultural connection, 1987. Chinese values and the search for culture-free dimensions of culture. J. Cross Cult. Psychol. 18 (2), 143–164.

12
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

Conner, M., Sparks, P., Povey, R., James, R., Shepherd, R., Armitage, C.J., 2002. Moderator effects of attitudinal ambivalence on attitude-behaviour relationships. Eur.
J. Soc. Psychol. 32 (5), 705–718.
Davis, F.D., 1989. Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Q. 13 (3), 319–340.
Dean, M., Raats, M.M., Shepherd, R., 2012. The role of self-identity, past behaviour and their interaction in predicting intention to purchase fresh and processed
organic food. J. Appl. Soc. Psychol. 42, 669–688.
Dodds, W.B., Monroe, K.B., Grewal, D., 1991. Effects of price, brand, and store information on buyers’ product evaluations. J. Mark. Res. 28 (3), 307–319.
Dunn, M.G., Murphy, P.E., Skelly, G.U., 1986. Research note: the influence of perceived risk on brand preference for supermarket products. J. Retail. 62 (2), 204–216.
Edison, S.W., Geissler, G.L., 2003. Measuring attitudes towards general technology: antecedents, hypotheses and scale development. J. Target. Meas. Anal. Mark. 12
(2), 137–156.
Featherman, M.S., Pavlou, P.A., 2003. Predicting e-services adoption: a perceived risk facets perspective. Int. J. Hum Comput Stud. 59 (4), 451–474.
Flamm, B.J., Agrawal, A.W., 2012. Constraints to green vehicle ownership: a focus group study. Transp. Res. Part D: Transp. Environ. 17 (2), 108–115.
Gärling, A., Thøgersen, J., 2001. Marketing of electric vehicles. Bus. Strateg. Environ. 10 (1), 53–65.
Globisch, J., Dütschke, E., Schleich, J., 2018. Acceptance of electric passenger cars in commercial fleets. Transp. Res. A Policy Pract. 116, 122–129.
Han, L., Wang, S., Zhao, D., Li, J., 2017. The intention to adopt electric vehicles: driven by functional and non-functional values. Transp. Res. A Policy Pract. 103,
185–197.
Hänze, M., 2001. Ambivalence, conflict, and decision making: attitudes and feelings in Germany towards NATO’s military intervention in the Kosovo war. Eur. J. Soc.
Psychol. 31 (6), 693–706.
Hass, R.G., Katz, I., Rizzo, N., Bailey, J., Moore, L., 1992. When racial ambivalence evokes negative affect, using a disguised measure of mood. Pers. Soc. Psychol. Bull.
18 (6), 786–797.
Hu, H.C., 1944. The Chinese concepts of face. Am. Anthropol. 46 (1), 45–64.
Hu, H., Hu, H., Parsa, H., 2017. Ambivalence attitudes toward green products and the moderating role of green advertisement: an extended abstract. In: Rossi, P. (Ed.),
Marketing at the Confluence Between Entertainment and Analytics. Springer, Cham, pp. 1289–1294.
Itzchakov, G., Van Harreveld, F., 2018. Feeling torn and fearing rue: Attitude ambivalence and anticipated regret as antecedents of biased information seeking. J. Exp.
Soc. Psychol. 75, 19–26.
Jaiswal, D., Kaushal, V., Kant, R., Kumar Singh, P., 2021. Consumer adoption intention for electric vehicles: Insights and evidence from Indian sustainable
transportation. Technol. Forecast. Soc. Chang. 173, 121089.
Jonas, K., Diehl, M., Brömer, P., 1997. Effects of attitudinal ambivalence on information processing and attitude-intention consistency. J. Exp. Soc. Psychol. 33 (2),
190–210.
Junquera, B., Moreno, B., Álvarez, R., 2016. Analyzing consumer attitudes towards electric vehicle purchasing intentions in Spain: technological limitations and
vehicle confidence. Technol. Forecast. Soc. Chang. 109, 6–14.
Kamal, S.A., Shafiq, M., Kakria, P., 2020. Investigating acceptance of telemedicine services through an extended technology acceptance model (TAM). Technol. Soc.
60, 101212.
Kamarudin, A., Naida, R., Basil, D., Nazzalyne, E.H., Joewono, P., Munzilah, R., Ezree, A.M., 2016. Driver intention to use electric cars using technology acceptance
model. J. Eng. Appl. Sci. 11, 1524–1528.
Kaplan, K.J., 1972. On the ambivalence-indifference problem in attitude theory and measurement: a suggested modification of the semantic differential technique.
Psychol. Bull. 77 (5), 361.
Kihm, A., Trommer, S., 2014. The new car market for electric vehicles and the potential for fuel substitution. Energy Policy 73, 147–157.
Kulkarni, G., Ratchford, B.T., Kannan, P.K., 2012. The impact of online and offline information sources on automobile choice behavior. J. Interact. Mark. 26 (3),
167–175.
Laroche, M., Yang, Z., McDougall, G.H.G., Bergeron, J., 2005. Internet versus bricks-and-mortar retailers: an investigation into intangibility and its consequences.
J. Retail. 81 (4), 251–267.
Le, A.N., Jain, M., Tam, T.D., Tam, L., 2019. Attitudinal ambivalence towards green products: an empirical study in an emerging market. Int. J. Econ. Business Res. 18
(3), 292–303.
Li, W., Long, R., Chen, H., Geng, J., 2017. Household factors and adopting intention of battery electric vehicles: a multi-group structural equation model analysis
among consumers in Jiangsu province, China. Natural Hazards 87 (2), 945–960.
Lim, M.K., Mak, H.Y., Rong, Y., 2015. Toward mass adoption of electric vehicles: impact of the range and resale anxieties. Manuf. Serv. Oper. Manage. 17 (1),
101–119.
Liu, Y., Shankar, V., 2015. The dynamic impact of product-harm crises on brand preference and advertising effectiveness: an empirical analysis of the automobile
industry. Manage. Sci. 61 (10), 2514–2535.
Mohr, L.A., Eroglu, D., Ellen, P.S., 1998. The development and testing of a measure of skepticism toward environmental claims in marketers’ communications.
J. Consum. Aff. 32 (1), 30–55.
Odekerken-Schröder, G., Hennig-Thurau, T., Knaevelsrud, A.B., 2010. Exploring the post-termination stage of consumer-brand relationships: an empirical
investigation of the premium car market. J. Retail. 86 (4), 372–385.
Olsen, S.O., Wilcox, J., Olsson, U., 2005. Consequences of ambivalence on satisfaction and loyalty. Psychol. Mark. 22 (3), 247–269.
Olsen, S.O., Prebensen, N., Larsen, T.A., 2009. Including ambivalence as a basis for benefit segmentation. Eur. J. Mark. 43 (5/6), 762–783.
Otnes, C., Lowrey, T., Shrum, L., 1997. Toward an understanding of consumer ambivalence. J. Consum. Res. 24 (1), 80–93.
Ou, X., Zhang, X., Chang, S., 2010. Scenario analysis on alternative fuel/vehicle for China’s future road transport: life-cycle energy demand and GHG emissions.
Energy Policy 38 (8), 3943–3956.
Pang, J., Keh, H.T., Li, X., Maheswaran, D., 2017. “Every coin has two sides”: The effects of dialectical thinking and attitudinal ambivalence on psychological
discomfort and consumer choice. J. Consum. Psychol. 27 (2), 218–230.
Park, E., Lim, J., Cho, Y., 2018. Understanding the emergence and social acceptance of electric vehicles as next-generation models for the automobile industry.
Sustainability 10 (3), 662–674.
Priester, J.R., Petty, R.E., 1996. The gradual threshold model of ambivalence: relating the positive and negative bases of attitudes to subjective ambivalence. J. Pers.
Soc. Psychol. 71 (3), 431.
Qian, L., Yin, J., 2017. Linking Chinese cultural values and the adoption of electric vehicles: the mediating role of ethical evaluation. Transp. Res. Part D: Transp.
Environ. 56, 175–188.
Qian, L., Grisolía, J.M., Soopramanien, D., 2019. The impact of service and government-policy attributes on consumer preferences for electric vehicles in China.
Transp. Res. A Policy Pract. 122, 70–84.
Rothman, N.B., Pratt, M.G., Rees, L., Vogus, T.J., 2017. Understanding the dual nature of ambivalence: why and when ambivalence leads to good and bad outcomes.
Acad. Manag. Ann. 11 (1), 33–72.
Russell, C.A., Russell, D.W., Klein, J., 2011. Ambivalence toward a country and consumers’ willingness to buy emblematic brands: The differential predictive validity
of objective and subjective ambivalence measures on behavior. Mark. Lett. 22 (4), 357–371.
Schuitema, G., Anable, J., Skippon, S., Kinnear, N., 2013. The role of instrumental, hedonic and symbolic attributes in the intention to adopt electric vehicles. Transp.
Res. A Policy Pract. 48, 39–49.
Su, C.-W., Yuan, X.i., Tao, R., Umar, M., 2021. Can new energy vehicles help to achieve carbon neutrality targets? J. Environ. Manage. 297, 113348.
Thompson, M.M., Zanna, M.P., Griffin, D.W., 1995. Let’s not be indifferent about (attitudinal) ambivalence. In: Petty, R.E., Krosnick, J.A. (Eds.), Attitude Strength:
Antecedents and Consequences. Erlbaum, Hillsdale, pp. 361–386.
Vahdat, A., Alizadeh, A., Quach, S., Hamelin, N., 2021. Would you like to shop via mobile app technology? The technology acceptance model, social factors and
purchase intention. Australas. Mark. J. 29 (2), 187–197.

13
L. Zhang et al. Transportation Research Part A 174 (2023) 103742

Wan, J., Zhao, L., Gupta, S., Lu, Y., 2017. Evaluating app bundling strategy for selling mobile apps: an ambivalent perspective. Inf. Technol. People 30 (1), 2–23.
Wang, S., Fan, J., Zhao, D., Yang, S., Fu, Y., 2016. Predicting consumers’ intention to adopt hybrid electric vehicles: using an extended version of the theory of planned
behavior model. Transportation 43 (1), 123–143.
Wang, S., Li, J., Zhao, D., 2017b. The impact of policy measures on consumer intention to adopt electric vehicles: Evidence from China. Transp. Res. A Policy Pract.
105, 14–26.
Wang, C.L., Lin, X., 2009. Migration of Chinese consumption values: traditions, modernization, and cultural renaissance. J. Bus. Ethics 88 (S3), 399–409.
Wang, N., Pan, H., Zheng, W., 2017a. Assessment of the incentives on electric vehicle promotion in China. Transp. Res. A Policy Pract. 101, 177–189.
Wang, Y.H., Wang, Q., 2013. Factors affecting Beijing residents’ buying behavior of new energy vehicle: an integration of technology acceptance model and theory of
planned behavior. Chinese J. Manage. Sci. 21 (S2), 691–698.
Wang, S., Wang, J., Li, J., Wang, J., Liang, L., 2018. Policy implications for promoting the adoption of electric vehicles: do consumer’s knowledge, perceived risk and
financial incentive policy matter? Transp. Res. A Policy Pract. 117, 58–69.
Wang, Y.u., Wang, S., Wang, J., Wei, J., Wang, C., 2020. An empirical study of consumers’ intention to use ride-sharing services: using an extended technology
acceptance model. Transportation 47 (1), 397–415.
Wang, J.M., Wu, L.C., 2015. Two-factor model of affection-behavior in green purchase: hypotheses and test. J. Manage. Sci. 24 (3), 258–270.
Wang, Z., Zhao, C., Yin, J., Zhang, B., 2017c. Purchasing intentions of Chinese citizens on new energy vehicles: how should one respond to current preferential policy?
J. Clean. Prod. 161, 1000–1010.
Watson, D., Clark, L., Tellegen, A., 1988. Development and validation of brief measures of positive and negative affect: the PANAS scales. J. Pers. Soc. Psychol. 54 (6),
1063–1070.
Wolff, S., Madlener, R., 2019. Driven by change: commercial drivers’ acceptance and efficiency perceptions of light-duty electric vehicle usage in Germany. Transport.
Res. Part C: Emerg. Technol. 105, 262–282.
Wong, N.Y., Ahuvia, A.C., 1998. Personal taste and family face: Luxury consumption in Confucian and western societies. Psychol. Mark. 15 (5), 423–441.
Wu, B., Chen, X., 2017. Continuance intention to use MOOCs: Integrating the technology acceptance model (TAM) and task technology fit (TTF) model. Comput. Hum.
Behav. 67, 221–232.
Wu, J., Liao, H., Wang, J.W., Chen, T., 2019. The role of environmental concern in the public acceptance of autonomous electric vehicles: a survey from China.
Transport. Res. F: Traffic Psychol. Behav. 60, 37–46.
Zhang, W., 2020. Consumption, taste, and the economic transition in modern China. Consum. Mark. Cult. 23 (1), 1–20.
Zhang, L., Qin, Q., 2018. China’s new energy vehicle policies: evolution, comparison and recommendation. Transp. Res. A Policy Pract. 110, 57–72.

14