Effect of Environmental Awareness on Purchase Intention and Satisfaction Pertaining to Electric Vehicles in Japan

Abstract

The reduction of CO2 emission using electric vehicles (EVs) is attracting much attention as a countermeasure for global warming. In this study, we investigate the intention of non-EV owners and the post-purchase satisfaction of EV owners by conducting online survey in Japan. The structural relation of both these factors is analyzed using structural equation modeling (SEM). This analysis focuses on the environmental awareness. We compare the estimations between the purchase intentions of non-EV users and the post-purchase satisfaction of EV users. Results show that the structures of purchase intentions of non-EV users and post-purchase satisfaction of EV are different. The evaluation of EVs shows that the environmental awareness has a direct effect on the purchase intention of a non-EV user, whereas an indirect effect on the post-purchase satisfaction of a EV user.

Full text

Effect of Environmental Awareness on Purchase Intention and Satisfaction
Pertaining to Electric Vehicles in Japan
Takanori OKADA1, Tetsuya TAMAKI2*, and Shunsuke MANAGI3
Abstract
The reduction of CO2 emission using electric vehicles (EVs) is attracting much attention
as a countermeasure for global warming. In this study, we investigate the intention of non-
EV owners and the post-purchase satisfaction of EV owners by conducting online survey
in Japan. The structural relation of both these factors is analyzed using structural equation
modeling (SEM). This analysis focuses on the environmental awareness. We compare the
estimations between the purchase intentions of non- EV users and the post-purchase
satisfaction of EV users. Results show that the structures of purchase intentions of non-
EV users and post-purchase satisfaction of EV are different. The evaluation of EVs shows
that the environmental awareness has a direct effect on the purchase intention of a non-
EV user, whereas an indirect effect on the post-purchase satisfaction of a EV user.
Keywords: SEM, CO2 reduction, EVs diffusion, purchase intention, post-purchase
satisfaction
Acknowledgements: The author would like to acknowledge the financial supports from
Specially Promoted Research through a Grant-in-Aid (Scientific Research 26000001)
from the Japanese Ministry of Education, Culture, Sports, Science and Technology
(MEXT) and the Environment Research and Technology Development Fund (S-16) from
the Japanese Ministry of the Environment.
1Graduate School of Engineering, Kyushu University, 744 Motooka Nishi-ku Fukuoka
819-0395, Japan
2Faculty of Engineering and Design, Kagawa University,2217-20 Hayashi-cho,
Takamatsu, Kagawa,761-0396, Japan
3Department of Urban and Environmental Engineering, Kyushu University, 744 Motooka
Nishi-ku Fukuoka 819-0395, Japan
*Corresponding Author: tamaki@eng.kagawa-u.ac.jp

1. Introduction
1.1 Climate change and the CO2 reduction target in Japan
Recently, various countries and international organizations have reported climate change
as an important problem. Greenhouse gases (GHGs) produced by human activities are the
primary factor for climate change. Atmospheric GHG concentration began to rise during
the industrial revolution and have risen considerably in recent decades (UNEP, 2012).
Total CO2 emissions from human activities from 1970 to 2011 account for half of the
total global emissions since the 1750s (IPCC, 2014), excluding natural volcanic and
oceanic CO2 emissions.
As one of the global warming countermeasures, advances in energy technology are
attracting considerable attention (Chu and Majumdar, 2012; Nykvist and Nilsson, 2015).
Japan’s Intended Nationally Determined Contributions policy pledged to reduce GHG
emissions by 26% by 2030 compared with those reported in 2013 at the COP21 held in
Paris, France in 2015. The Japanese government also announced that it will increase its
spending by 1.3 trillion yen each year, starting in 2020, to support the public and private
sectors in developing countries focus on climate change and promote innovative
technology developments. Furthermore, the Japanese government set out the Japan
Revitalization Strategy and said that Japan aims to increase the percentage of next-
generation vehicles (NGVs) to be 50%–70% of new car sales by 2030, promoting the
creation of initial demand, supporting research and development on vehicle performance
improvement, and fomenting efficient infrastructure development (Prime Minister of
Japan and His Cabinet, 2013).
Since ratifying the Kyoto Protocol in 2005, Japan has implemented CO2 emission
reduction strategies in the industrial, service, transportation, and energy conversion
sectors, which have achieved some positive results. In particular, NGVs have been
adopted in the transportation sector. NGVs, including EVs, consume less fuel, emit less
GHGs, and are expected to have a considerable effect in reducing CO2 emissions. EVs
are also expected to be used as emergency power supply for disasters and electric outages.
In Japan, CO2 emissions from the transportation sector account for ~19% of total
emissions and emissions from automobiles account for 87.8% (MLIT, 2013a). In addition,
the number of domestic EVs has grown from ~9,000 in 2010 to ~70,000 in 2014. The
domestic number continues to increase; however, EVs currently account for <1% of
automobiles, whereas hybrid vehicles account for ~20% (International Energy Agency,
2017). Thus, Japan has the potential to reduce CO2 emissions by expanding NGV
ownership. Some researchers suggest that there will be negative outcomes from EV
diffusion policies (Massiani, 2015; Kazimi, 1997; Funk and Rabl, 1999). However, Ito,

et al. (2013) insist that the infrastructure investments of EV become efficient once EV’s
share exceeds 5.63% of annual new vehicle sales. Ito and Managi (2015) show the
potential impact of EV diffusion using cost-benefit analysis. Their study points out that
one obstacle to overcome has been the electric car battery and major technological
progress is required to reduce its production costs and improve EV performance. Doing
so would increase the adoption of EVs.
1.2 Literature reviews
NGVs, including EVs, are automobiles that excel in environmental performance with
little or no emission of air pollutants. Many countries have attempted to diffuse EVs,
however, the trends of the sales volumes vary from country to country. The U.S. could
not achieve the goal of placing one million PEVs (plug-in electric vehicles) by 2015
(Shepardson and Woodall, 2016). On the other hand, Norway is a global forerunner in the
field of EVs. One of the largest reasons is the exemption from sales tax offered, which
was shown to be a critical incentive for more than 80% of respondents (Bjerkin et al.,
2016). China, also, is rapidly increasing the market share of EVs (Xu and Su, 2016).
Besides these countries, many researchers investigate the EV adoption incentives in each
country such as US (Hidrue et al., 2011; Tanaka et al., 2014; Flamm, 2009), Canada
(Daziano, 2012), UK (Lane and Potter, 2007), Netherlands (Brownstone et al., 2014),
Malaysia (Adnan et al., 2017), and so forth. Some factors would be common to all, others
would depend on the national traits. Rezvani, et al. (2015) reviewed the studies related to
consumer EV adoption based on individual-specific psychological factors. Liao, et al.
(2017) complemented the overview and reviewed the studies applying economic
approaches. Liao, et al. (2015) insisted that financial, technical and infrastructure
attributes have a significant impact on EV choice. Coffman, et al. (2017) also identified
fifty studies related to EV adoption and organized them by the factor(s) that affected
adoption.
Several approaches have been used to analyze the diffusion of EVs into the market and
various factors related to adopting EV have been analyzed by many researchers. Some
researchers have estimated discrete choice models based on stated preference (SP) data.
Caulfield, et al. (2010) investigated factors influencing individual decision-making using
a multinomial Logit (MNL) model. An MNL model is one of several basic analysis
models (e.g. Brownstone et al., 2000), however it assumes independence from irrelevant
alternatives (IIA). Thus, some studies use a nested Logit model and a mixed Logit one to
relax the IIA (Potoglou and Kanaroglou, 2007). Other models are also used in the analysis.
Hidrue, et al. (2011) calculated car owners’ willingness to pay for vehicle attributes using

a random utility model. Li, et al. (2012) reported on factors that affect the purchase
intentions of flexible-fuel vehicles and hybrid vehicles using a Probit model. Jabbari, et
al. (2017) used a Wilcoxon signed rank test, a McNemar test and a Chi-squared test to
determine the reasons for rejecting the vehicles under consideration. In addition, some
researchers also analyzed outcomes by structural equation modeling (SEM) (Degirmenci
and Breitner, 2017; Nayum and Klockner, 2014). One of the beneficial features of SEM
is being able to assess the relationships between factors. Human behavior has been
modeled by some researchers. Ajzen, (1991) suggested the theory of planned behavior
(TPB), which is a theory explaining intentional action. TPB assumes humans choose their
behavior based on rational evaluations. One more famous model is the value-belief-norm
(VBN) theory (Stern, 2000). A SEM can describe the relationship between the factors
based on these theories.
Some forecasts of EV market diffusion are predicted by forecasting models (Becker
et al., 2009; Al-Alawi & Bradley, 2013). These models are based on the Bass model (Bass,
1969), which simulates a diffusion process of a new production, particularly of durable
consumer goods. Simulation models have also been used to forecast and evaluate the
diffusion of EVs into the market (Massiani, 2015; Greene, 2001; Santini and Vyas, 2005;
AECOM Australia, 2009). These models tend to provide a wider approach to the diffusion
pattern compared with the previous studies. In addition, they tend to estimate market
diffusion by considering economic scenarios.
This study analyzes market diffusion using SP data. Although forecasting models and
simulation models are important tools for evaluating the diffusion of EVs, this study aims
to outline the factors that promote the spread of EVs without focusing on economic
impacts. Thus, we clarify purchase factors and their relationships, using SEM, and
propose measures toward EV market share expansion. We focus on EVs that have already
been put to practical use as NGVs and clarify purchasing factors and their relationships.
In particular, we focus on the effects related to functional aspects, such as vehicle
performance, and general environmental awareness (e.g., ecosystem conservation and
product reuse). Environmental awareness is considered in several studies. Some of them
show that environmental awareness has an effect on EV adoption (Flamm et al., 2009;
Carley et al., 2013), while others could not find any obvious effect (Zhang et al., 2011).
Although Noppers, et al. (2014) clarified how important environmental attributes
compared with other (instrumental or symbolic) attributes when they were analyzed, the
environmental awareness referred to these studies has to do with respondents’ sense about
global warming, especially CO2 emission reduction. Thus, environmental awareness is
not precisely measured. In addition, Nopper, et al. (2015) considered consumers’ adoption

stage based on the theory of Rogers (1962), which suggested classifying consumers into
the segments based on the adoption timing. The studies distinguishing the EV adaptation
timing like Nopper, et al. (2015) are important, but little is known about the differences.
In particular, it is difficult to classify the adoption timing of things in the early stages of
the diffusion, such as EV. If we follow the theory of Rogers (1962), consumers are divided
into five adoption segments; innovators, early adopters, early majority, late majority, and
traditionalists. Innovators means those who are first to adopt an innovative product (the
top 2.5 % of the group adopting the innovation), but EVs currently account for <1% of
automobiles in Japan. Hence, this study clarifies that the difference between the people
who do not have EV and the people who still do. In addition, as mentioned above, the
situation of EV diffusion will differ from country to country and depend on national traits
or policies. We focus on Japanese EV markets. Tanaka, et al. (2014) estimated the
willingness to pay for EVs in the U.S. and Japan, and showed different factors between
the countries, but that study did not focus on the relationship between factors.
This remainder of this paper is organized as follows. Section 1 reviews the current
literature, Section 2 clarifies the scope of this work and introduces the dataset, Section 3
show the SEM and the results and Section 4 discuss the results. Finally, conclusions and
future study are summarized in Section 5.
2. Data
We conducted a nationwide online survey in Japan between November 16 and
December 14, 2015. There were 246,642 respondents; 785 of whom were EV owners.
The survey is conducted for all generations, and consist of questionnaires related to
lifestyle, happiness, or satisfaction, including topics about EVs
1
. This study uses the
questions related to EVs and demographics except for the other questions that are not
relevant for this study. Before the large-scale survey started, a pre-survey was carried out
for tuning the questionnaires.
[Table 1]
Although EV users account for <1% of total car users, this survey could obtain
sufficient data to analyze and compare the purchase intentions of non-EV users and the
post-purchase satisfaction of EV users (all questions can be found in the Appendix). This
1
We requested an outside company (Nikkei Research Inc.).

survey considers environmental awareness along with the environmental performance of
EVs and the awareness of EV owners toward environmental policies. By representing
environmental awareness as the awareness toward environmental policies, we focus on
the impact on the purchase intentions of people who hope changing our environment by
policies. Therefore, the environmental awareness in our study includes general
environmental awareness. We classified environmental policy into eight items based on
previous studies (House of Councilors, The National Diet of Japan, 2015). The
participants were questioned, and their responses were evaluated as the propriety on a
scale from one to five. Here, we evaluate their stance on renewable energies, NGVs,
environmental conservation, air and water pollution, conservation of endangered species,
reuse and recycling, garbage disposal, and CO2 with questions such as, “How important
is the policy to you?” The scale of responses is as follows: 1 for very unimportant; 2 for
unimportant; 3 for neither important nor unimportant; 4 for important; 5 for very
important.
In addition, the survey included several questions about vehicles (see Appendixes 3)
based on previous studies (Ministry of Economy, Trade and Industry, 2009; Next
Generation Vehicle Promotion Center, 2012). With respect to EV purchase intentions, we
asked “What are the merits of purchasing EVs? Please choose all the following choices
that are applicable to you.” We suggested fifteen choices, including “I do not know,” and
obtained information about the respondents’ usual means of transportation, the type of
car, and the average length of a journey. We evaluated non-EV participants’ EV purchase
intentions and the post-purchase satisfaction of EV users. Non-EV users’ responses
belonged to the following choices: i) I do not know; ii) I have not considered purchasing
a car; iii) I am interested in EVs but have not considered purchasing it for some time; iv)
I have considered purchasing a car but not an EV; and v) I have considered purchasing an
EV. We use the data as a binary variable; 0 is assigned to responses ii), iii) and iv), and 1
is assigned to response v). EV users were asked about their satisfaction level with EVs
and could choose the following responses: 5 Completely satisfied; 4 Slightly satisfied; 3
Neither; 2 Slightly dissatisfied; 1 Completely dissatisfied (Appendix 4).
In addition, this study made some demographic observations, such as gender, age, and
annual household income. During the analysis, we excluded the responses “I don’t know”
about the purchase intentions of EVs, “I don’t know/I don’t want to answer” about the
annual household income, and “I don’t know” about the importance of environmental
policy. We also excluded the responses of those who do not own a car. Thus, the number
of usable responses was reduced to 106,446 car owners and 536 EV owners. Table 1
summarized the demographic statistics of the respondents. The high proportion of older

people is one factor that limited the number of automobile owners. EV users have higher
income than non-EV users, typically around six million yen. For both groups, ~70% of
journeys take less than one hour and 20% take between one and three hours. 60% of non-
EV users own only one car, whereas most EV users (70%) have more than one car.
3 Models and results
3.1 Structural Equation Modeling
SEM is performed to assess the relationship between factors considering unobservable
variables. Herein, the samples are classified as EV owners and non-EV owners; models
of EV owners post-purchase satisfaction and purchase intentions for non-EV owners are
built. The purchase intentions of respondents who do not own an EV is shown (Model I)
in Figure 1, and the post-purchase satisfaction of EVs owners (Model II) is shown in
Figure 2. The path coefficients shown in these figures are standardized solutions. In
addition, the solutions are shown in appendices 5 and 6 and we evaluated four goodness-
of-fit indices such as GFI (Goodness of Fit Index), AGFI (Adjusted Goodness of Fit
Index), CFI (Comparative Fit Index), and RMSEA (Root Mean Square Error of
Approximation).
2
We used STATA 15 for the analysis.
We set two latent variables such as “environmental awareness” and “evaluation of EVs”
in the models. In terms of environmental awareness, the latent variable is defined by the
observable variables that are related to the stances on several environmental policies (see
Appendixes 1 and 2). Cronbach’s alpha
3
of the 8 items in Model I and Model II became
0.902 and 0.912, respectively. As for evaluating EVs, this latent variable is based on
previous studies (Ministry of Economy, Trade and Industry, 2009; Next Generation
Vehicle Promotion Center, 2012; Ministry of Economy, Trade and Industry, 2011) (see
Appendix 3). Our models are based on the model constructed by Degirmenci and Breitner
(2017); purchase intentions are explained by attitudes toward EVs and the attitude is
explained by the factors related to EVs and individual attributions. However, that model
did not analyze the consumer’s general environmental awareness and focused on
functional factors such as vehicle performance and price. This study focuses on
environmental awareness instead of simplifying consumer attitudes. Environmental
2
GFI and AGIF are the indexes, which stand for the persuasiveness of models. The values are
between 0 and 1, and are preferable that the value are closer to 1. In addition, it is better that the gap
between GFI and AGIF is small. CFI is not very sensitive to sample size and the acceptable range is
> 0.90. RMSEA is currently the most popular measure of model fit. The acceptable range is < 0.08.
3
Cronbach’s alpha reflects a reliable scale of items. The acceptable range is > 0.6.

awareness not only affects purchase intentions (Carley et al., 2013; Zhang et al., 2011;
Flamm, 2009), but also imputes several factors such as gender and age (Szagun and
Pavlov, 1995). The demographics is used as control variables with reference of
Degirmenci and Breitner (2017). This study constructs models that consider these effects.
3.2 Model I (non-EV owners)
We verified the relation between observed and latent variables for non-EV users in
Model I. The GFI, AGFI, CFI, and RMSEA are 0.915, 0.894, 0.864, and 0.055,
respectively. These values are generally within or near each variable’s acceptable range.
This analysis uses quite a large sample so that CFI and RMSEA have the potential to fall
outside, but GFI and AGFI are not bad outcomes.
Results show that environmental awareness (

= 0.038; p < 0.01) and the evaluation
of EVs (

= 0.087; p < 0.01) directly influenced purchase intentions. In addition, the
evaluation of EVs was positively influenced by environmental awareness (

= 0.287; p <
0.01). Hence, environmental awareness has direct and indirect effects, which is seen via
the evaluation of EVs on purchase intentions. On the one hand, the direct effect on
environmental awareness to purchase intention has a coefficient of 0.038, on the other
hand the indirect effect is 0.025, which can calculate by multiplying the coefficient of the
path from environmental awareness to evaluation of EVs by the coefficient of the path
from evaluation of EVs to purchase intention (Bollen, 1987). Therefore, the direct effects
are one and a half times larger than indirect effects. Regarding respondent demography,
age significantly affects environmental awareness and evaluation of EVs. Gender is also
significant, and data show that men have a higher evaluation of EVs but less
environmental awareness. Household income has no significant influence on
environmental awareness. In contrast, data on the drive time can explain environmental
awareness.
4
[Figure 1]
4
In this study, we assumed that the people who do not have EV purchase intention mean the people
who do not have an interest in EVs. On the other hand, we also analyzed the case in which the purchase
intentions variable was classified; 0 is assigned to responses iii) and iv), and 1 is assigned to response
v). The estimated coefficients are 0.050 (environmental awareness -> purchase intentions; p< 0.01),
0.045 (evaluation of EVs -> purchase intentions; p< 0.01) and 0.274 (environmental awareness ->
evaluation of EVs; p< 0.01). Although the value from environmental awareness to evaluation of EVs
became smaller, there is no big difference between them.

3.3 Model II (EV owners)
Model II shows the relation between observable and latent variables, including the level
of post-purchase satisfaction for EV owners. Goodness-of-fit indices, such as GFI, AGFI,
CFI, and RMSEA are 0.858, 0.824, 0.867, and 0.061, respectively. These values are
generally within or near each variable’s acceptable range.
Model II focuses on environmental awareness; environmental awareness (

= 0.29; p
< 0.01) directly influenced evaluation of EVs; however, the direct effect of environmental
awareness on the level of satisfaction is insignificant. In Model I, the direct effect from
environmental awareness and the indirect effect via the evaluation of EVs are observed.
However, in Model II, only the evaluation of EVs (

= 0.28; p < 0.01) is a regulatory
factor for the level of satisfaction and other factors influence the level of satisfaction via
the evaluation of EVs. Age has a significant influence on both environmental awareness
and the evaluation of EVs, similar to Model I. However, gender is only significant with
respect to environmental awareness and household income is only significant with regard
to the evaluation of EVs. There are also differences between the two models in the amount
of influence from normal use patterns. In Model II, only drive time has a positive
significant effect on the evaluation of EVs.
[Figure 2]
4 Discussion
Model I represents the structures of the purchase intention of non-EV owners, and
Model II represents the post-purchase satisfaction of EV owners. Both environmental
awareness and the evaluation of EVs have significant influences on purchase intentions
in Model I. It means that improving the evaluation of EVs and environmental awareness
would lead to increased purchase intentions. Although evaluation of EVs has a stronger
influence than environmental awareness, environment awareness also explains the
evaluation of EVs. In contrast, environmental awareness does not have significant direct
influence on post-purchase satisfaction in Model II. On the other hand, environmental
awareness has significant influence on the evaluation of EVs and the coefficient is high
compared with Model I. High environmental awareness might have the potential to
indirectly arouse higher satisfaction by becoming a cause of the evaluation of EVs. Hence,
it would be important to raise environmental awareness so as to enhance the evaluation
of EVs. In our analysis, even though we found a significant influence from environmental
awareness and the evaluation of EVs on purchase intentions in Model I, the effects are

small. Noppers, et al. (2014) insisted that the environmental attribute is important, and
people are motivated to adopt sustainable innovations, such as EVs, because of their
environmental benefits. Furthermore, Carley, et al. (2013) estimated that an
environmental views index becomes significant, while the effect size may not be very
large. As with Carley, et al. (2013), in our study, environmental awareness contains factors
other than those related to EV diffusion effects, such as global warming. It would have
the potential to become a small effect on purchase intentions. Yet, on the contrary, the
effect size from the environmental awareness to the evaluation of EVs is not small. We
found that the general environmental awareness (as we use it) has a small impact on
purchase intentions and satisfaction but the awareness does have sufficient impact on the
evaluation of EVs. Environmental awareness not directly related to EV diffusion effects
is an important factor in the evaluation of EVs. A following point can be given as the
other reason for the small effect. We use binary variable as the purchase intention index.
People who have purchase intentions is only 5.5 % and they may be people who seriously
considering the purchase. By Carley et al. (2013), 3.5% of respondents indicate serious
consideration of buying PEV. Those who lightly have an interest in EV might be included
in remaining 94.5%. Hence, the effect size from the environmental awareness to the
evaluation of EVs might be small. Governments or institutions hoping to promote EVs
should publish detailed information not only about EV performance but also the huge
positive impact EVs have on the natural environment, e.g., by recognizing the amount of
CO2 emissions reduced by the use of EVs. In addition, promoting the worth of natural
environment might be effective in evaluating EVs. Note that the environmental awareness
in this study consists in thinking about environmental policies. The fact that it becomes
high might indicate that people hope more specific environmental improvement behavior.
It is also important to compare the difference between the result using the environmental
awareness we defined and the one defined by other researchers.
Age is a control variable that affects both environmental awareness and the evaluation
of EVs in both models. This means that people become more interested in the
environment and have a better evaluation of EVs as they get older, depending on whether
age becomes a significant variable (Degirmenci and Breitner, 2017), or if the effect is
positive (Zhang et al, 2011) or negative (Hidrue et al., 2011; Ziegler, 2012). Gender and
household income also depend on the study shown by Liao, et al. (2017). In our study,
gender has significant negative influence on the environmental awareness (in both
models) and a positive influence on evaluation of EVs (in Model I). Household income
only has negative influence on the evaluation of EVs (in Model II). Promotions
advertising an EV’s environmental impact would be more effective in increasing female

purchase intentions, whereas advertisements focusing on performance would be more
effective on males. Furthermore, the number of vehicles is significantly related to
environmental awareness and the evaluation of EVs in Model I. Hidrue, et al. (2011)
stated that multicar households reduced the likelihood of being in the EV class, but our
result supports Jensen, et al. (2013) and the assumption that a multicar household would
not be constrained by limited drive range and would thus be prone to adopt EVs (Kurani
et al, 1996). Furthermore, EV owners’ satisfaction from EVs may not depend on the
number of own cars. In addition, we determine that drive time, which is one of the control
variables, negatively explains the evaluation of EVs. Driving range is one of the most
critical factors pertaining to EVs (Jensen et al., 2013). The variables related to the driving
range is included in almost EV adoption studies. It is one of the critical issues, but the
important level would change by sustainable innovations, e.g., charging station density,
charging time reductions (Dimitropoulos et al., 2013). Based on our results, promoting
EVs to older people and those who use vehicles mostly to go to nearby places would
effectively impact EV diffusion. Japan’s aging rate has exceeded 20% and young people
tend to shy away from driving (MLIT, 2013b). The market for middle- and older-age
Japanese drivers is comparatively large, such that certain results of EV diffusion would
be prospective. In addition, consumers possibly prefer using EVs to travel to nearby
places. More than half of drivers in Japan use cars to travel 20 kilometers every weekday
(Yabe et al. 2011) and our survey shows that approximately 80% of drivers use their car
for three hours per day. The consumers corresponding to this framework would be open
to considering EVs without much anxiety.
Although there are some differences in the significance of factors between EVs owners
and non-EV ones, some factors affect purchase intentions and post-purchase satisfaction
in both models. In other words, if both can be enhanced, the possibility of repurchasing,
and of additional purchases, will also expand and the NGVs ownership base will increase.
Increasing not only the purchase intentions but also the post-purchase satisfaction is
important for expanding EV ownership. Kuo, et al. (2009) showed that satisfaction and
post-purchase intention share a positive relation. Although the study targets mobile value-
added services, consumers would tend to repeat the service if they are satisfied even if it
is not a mobile value-added service. Then, the average upgrade cycle of vehicles is long
(9.3 years in Japan, according to the Cabinet Office Government of Japan, 2018) but
consumers must repurchase EVs for sustained diffusion. Consequently, promotions
should be inclined toward both EV and non-EV owners. Some promotions are predictably
effective for both types of owners, but many have a potential for only one type. Thus, the
promotion strategy must be changed depending on the diffusion situation.

Finally, we consider the limitations of our study. Goodness-of-fit indices of our models
are generally within or near each variable’s acceptable range, however they might be not
good scores considering large number of sample size. It might be improved by collecting
many variables and recreating by more complex models. For example, the factor of
vehicle type or vehicle price is not included in our models. There are several kinds of EVs
such as sports cars and vans, and those factors have the potential to arise the difference
of purchase intention or the difference of satisfaction. If we can distinguish them well,
the models can make more accurate estimates. This study also considers only results from
Japan, and the results may vary if the same survey is conducted in another country. Results
will likely depend on factors such as nationality, policies, and infrastructure. However,
transmitting knowledge from each country to future researchers is always important.
Accordingly, it is appropriate for studies to have a national scope because most policy
changes begin at the national level first. Moreover, although we gather large sample data,
we cannot represent sub-sample of Japanese population and we cannot collect enough
variables to know why EV owners might not be satisfied or how to improve satisfaction
and thus increase EV acceptance. In addition, SEM can decide which relationships
between factors do not contradict the data, but cannot declare that a single factor is the
main cause of a relational factor. We chose model II compared with the model showing
contrary relationships between evaluation of EVs and satisfaction. Model II has a better
fit than the other models by comparison. However, we cannot determine the causation of
any particular factors. There are methods with the instrumental variables (IV) or the
generalized method of moments (GMM) to estimate highly accurate causations. However,
it is not easy to find a relevant IV. Nevertheless, that is one difficulty of causality analysis
in general; if we can find the IV, the analysis would render more valuable estimates.
5. Conclusion
In this study, we conducted a nationwide online survey related to the purchase
intentions and post-purchase satisfaction of EVs in Japan. The total number of
respondents was 246,642, of which 785 were EV owners. We clarified the relation using
SEM. We built and analyzed two models, which showed the purchase intentions of non-
EV owners (Model I) and the post-purchase satisfaction of EV owners (Model II). In
Model I, a positive pass was observed from environmental awareness to the evaluation of
EVs and environmental awareness was clarified as having had a direct effect on purchase
intentions, whereas the evaluation of EVs had an indirect effect on it. In addition, the
evaluation of EVs was a stronger factor than environmental awareness for explaining or
predicting purchase intentions. In Model II, post-purchase satisfaction was not directly

explained by environmental awareness. To aim for a sustainable expansion of EV
ownership, improving purchase intentions and post-purchase satisfaction are important
elements. EVs have potential for global reduction of CO2, and carefully considered
promotion strategies are, therefore, important.
Reference
Adnan, N., Nordin, S. M., & Rahman, I., 2017. Adoption of PHEV/EV in Malaysia: a critical review
on predicting consumer behaviour. Renewable and Sustainable Energy Reviews, 72, 849-862.
AECOM Australia, 2009. Economic Viability of Electric Vehicles, for the Department of Environment
and Climate Change: 88 p.
Ajzen, I., 1991. The theory of planned behavior. Organizational behavior and human decision
processes, 50(2), 179-211.
Al-Alawi, B. M., & Bradley, T. H., 2013. Review of hybrid, plug-in hybrid, and electric vehicle market
modeling studies. Renewable and Sustainable Energy Reviews, 21, 190-203.
Bass, F. M. 1969. A new product growth for model consumer durables. Management science, 15(5),
215-227.
Becker, T. A., Sidhu, I., & Tenderich, B, 2009. Electric vehicles in the United States: a new model
with forecasts to 2030. Center for Entrepreneurship and Technology, University of California,
Berkeley, 24.
Bjerkan, K. Y., Nørbech, T. E., & Nordtømme, M. E., 2016. Incentives for promoting battery electric
vehicle (BEV) adoption in Norway. Transportation Research Part D: Transport and Environment, 43,
169-180.
Bočkarjova, M., Rietveld, P., Knockaert, J., & Steg, L., 2014. Dynamic consumer heterogeneity in
electric vehicle adoption. innovation, 3, 4.
Bollen, K. A., 1987. Total, direct, and indirect effects in structural equation models. Sociological
methodology, 37-69.

Brownstone David, Bunch S. David, Train Kenneth, 2000. Joint mixed logit models of stated and
revealed preferences for alternative-fuel vehicles. Transportation Research Part B 34, 315-338
Cabinet Office Government of Japan, 2018. Consumer confidence survey
http://www.esri.cao.go.jp/jp/stat/shouhi/honbun.pdf (accessed 4/17/2018)
Carley Sanya, Krause M. Rachel, Lane W. Bradley, Graham D. John, 2013. Intent to purchase a plug-
in electric vehicle: A survey of early impressions in large US cites, Transportation Research Part D 18,
39-45
Caulfield Brian, Farrell Seona, McMahon Brian, 2010. Examining individuals preferences for hybrid
electric and alternatively fuelled vehicles. Transportation Policy 17, 381-387
Chu, S., and Majumdar, A., 2012. Opportunities and challenges for a sustainable energy future. nature,
488(7411), 294.
Coffman, M., Bernstein, P., and Wee, S. 2017. Electric vehicles revisited: a review of factors that affect
adoption. Transport Reviews, 37(1), 79-93.
Daziano, R. A., 2012. Taking account of the role of safety on vehicle choice using a new generation
of discrete choice models. Safety Science, 50(1), 103-112.
Degirmenci Kenan and Breitner H. Michael, 2017. Consumer purchase intentions for electric vehicles:
Is green more important than price and range? Transportation Research Part D 51, 250-260
Dimitropoulos, A., Rietveld, P., & Van Ommeren, J. N., 2013. Consumer valuation of changes in
driving range: A meta-analysis. Transportation Research Part A: Policy and Practice, 55, 27-45.
Flamm Bradley, 2009. The impacts of environmental knowledge and attitudes on vehicle ownership
and use. Transportation Research Part D 14, 272-279
Funk, K., & Rabl, A., 1999. Electric versus conventional vehicles: social costs and benefits in France.
Transportation Research Part D: Transport and Environment, 4(6), 397-411.
Greene, D.L., 2001. TAFV Alternative Fuels and Vehicles Choice Model Documentation., Center for
Transportation Analysis. Oak Ridge National Laboratory, Oak Ridge, TN.

Hidrue K. Michael, Parsons R. George, Kempton Willett, Gardner P. Meryl, 2011. Willingness to pay
for electric vehicles and their attributes. Resource and Energy Economics 33, 686-705
House of Councillors, The National Diet of Japan, 2015. Chikyūkankyōmondai to senshinkoku no
kankyō seisaku (Global environmental problem and environmental policy of developed country)
Intergovernmental Panel on Climate Change, 2014. Climate Change 2014 Synthesis Report
International Energy Agency, 2017. Global EV Outlook 2017
Ito, Y., & Managi, S., 2015. The potential of alternative fuel vehicles: A cost-benefit analysis. Research
in Transportation Economics, 50, 39-50.
Ito, N., Takeuchi, K., & Managi, S., 2013. Willingness-to-pay for infrastructure investments for
alternative fuel vehicles. Transportation Research Part D: Transport and Environment, 18, 1-8.
Jabbari, P., Chernicoff, W., & MacKenzie, D., 2017. Analysis of Electric Vehicle Purchaser
Satisfaction and Rejection Reasons. Transportation Research Record: Journal of the Transportation
Research Board, (2628), 110-119.
Jensen, A. F., Cherchi, E., & Mabit, S. L., 2013. On the stability of preferences and attitudes before
and after experiencing an electric vehicle. Transportation Research Part D: Transport and Environment,
25, 24-32.
Kazimi, C., 1997. Valuing alternative-fuel vehicles in Southern California. The American Economic
Review, 87(2), 265-271.
Kuo Ying-Feng, Chi-Ming Wu, Wei-Jaw Deng, 2009. The relationships among service quality,
perceived value, customer satisfaction, and post-purchase intention in mobile value-added services.
Computers in Human Behavior, Volume 25 Issue 4, 887-896
Kurani, K. S., Turrentine, T., & Sperling, D., 1996. Testing electric vehicle demand in ‘hybrid

households’ using a reflexive survey. Transportation Research Part D: Transport and Environment,
1(2), 131-150.
Lane, B., & Potter, S., 2007. The adoption of cleaner vehicles in the UK: exploring the consumer
attitude–action gap. Journal of cleaner production, 15(11-12), 1085-1092.
Li Xiaogu, Clark D. Christopher, Jensen L. Kimberly, Yen T. Steven, English C. Burton, 2013.
Consumer purchase intentions for flexible-fuel and hybrid-electric vehicles. Transportation Research
Part D 18, 9-15
Liao, F., Molin, E., & van Wee, B., 2017. Consumer preferences for electric vehicles: a literature
review. Transport Reviews, 37(3), 252-275.
Massiani, J., 2015. Cost-Benefit Analysis of policies for the development of electric vehicles in
Germany: Methods and results. Transport policy, 38, 19-26.
Ministry of Economy, Trade and Industry, 2009. Fumin shijō-sha-muke EV PHV-tō ni kansuru ankēto
chōsa (Questionnaire survey on EV·PHV etc. for Kyoto citizens test drivers)
Ministry of Economy, Trade and Industry, 2011. Jisedai jidōsha (EV) o meguru yūza no ninshiki
(Recognition of the user over next-generation vehicles (EVs))
Ministry of Land, Infrastructure, Transport and Tourism, 2013a. Un'yu bumon ni okeru
chikyūondankataisaku (global warming countermeasures in the transportation sector) (in japanese)
Ministry of Land, Infrastructure, Transport and Tourism, 2013b. white paper on land, infrastructure,
transport and tourism in japan (in japanese)
Nayum, A., & Klöckner, C. A., 2014. A comprehensive socio-psychological approach to car type
choice. Journal of Environmental Psychology, 40, 401-411.
Next Generation Vehicle Promotion Center, 2012. Denki jidōsha jūden infura-tō no fukyū ni kansuru
chōsa (Survey on the spread of electric vehicles·charging infrastructure etc.)
Noppers, E. H., Keizer, K., Bolderdijk, J. W., & Steg, L., 2014. The adoption of sustainable

innovations: driven by symbolic and environmental motives. Global Environmental Change, 25, 52-
62.
Noppers, E. H., Keizer, K., Bockarjova, M., & Steg, L., 2015. The adoption of sustainable innovations:
The role of instrumental, environmental, and symbolic attributes for earlier and later adopters. Journal
of environmental psychology, 44, 74-84.
Nykvist, B., & Nilsson, M., 2015. Rapidly falling costs of battery packs for electric vehicles. nature
climate change, 5(4), 329.
Potoglou, D., & Kanaroglou, P. S., 2007. Household demand and willingness to pay for clean vehicles.
Transportation Research Part D: Transport and Environment, 12(4), 264-274.
Prime Minister of Japan and His Cabinet, 2013. Japan Revitalization Strategy
Rezvani, Z., Jansson, J., & Bodin, J., 2015. Advances in consumer electric vehicle adoption research:
A review and research agenda. Transportation research part D: transport and environment, 34, 122-
136.
Rogers, E. M., 1962. Diffusion of innovations. New York: The Free Press.
Santini, D.J., Vyas, A.D., 2005. Suggestions for a New Vehicle Choice Model Simulating Advanced
Vehicles Introduction Decisions (AVID): Structure and Coefficients. The University of Chicago,
Argonne National Laboratory, Oak Ridge, TN.
Shepardson, D., and Woodall, B., 2016. Electric vehicle sales fall far short of Obama goal. Reuter,
https://www.reuters.com/article/us-autos-electric-obama-insight-idUSKCN0UY0F0
Stern, P. C., 2000. New environmental theories: toward a coherent theory of environmentally
significant behavior. Journal of social issues, 56(3), 407-424.
Szagun Gisela, Pavlov I. Vladimir, 1995. Environmental awareness: A comparative study of German
and Russian adolescents. Youth & Society, Volume 27, n1 p93-112
Tanaka, M., Ida, T., Murakami, K., & Friedman, L., 2014. Consumers’ willingness to pay for
alternative fuel vehicles: A comparative discrete choice analysis between the US and Japan.

Transportation Research Part A: Policy and Practice, 70, 194-209.
United Nations Environment Programme, 2012. Global Environment Outlook-5
Xu, L., & Su, J., 2016. From government to market and from producer to consumer: Transition of
policy mix towards clean mobility in China. Energy Policy, 96, 328-340.
Yabe, K., Tokami, T., Shinoda, Y., Seki, T., Tanaka, H., and Akisawa, A., 2011. The Distribution of
Daily Drive Length and CO2 Reduction Effect by PHEVs, Journal of Japan Society of Energy and
Resources, Vol. 32, No. 4, 18-22. (in Japanese)
Zhang Yong, Yu Yifeng, Zou Bai, 2011. Analyzing public awareness and acceptance of alternative fuel
vehicles in China: The case of EV. Energy Policy 39, 7015-7024
Ziegler, A., 2012. Individual characteristics and stated preferences for alternative energy sources and
propulsion technologies in vehicles: A discrete choice analysis for Germany. Transportation Research
Part A: Policy and Practice, 46(8), 1372-1385.

Table 1
Table 1 Demographic statistics of respondents
Individual attributes, Actual use of
automobiles (%)
Total
(N = 246,642)
Model I
Non-EV user
(N = 106, 446)
Model II
EV user
(N = 536)
Gender
Male
59.00
70.70
69.59
Female
41.00
29.30
30.41
Age
19 years or below
0.32
0.12
0.93
20s
6.55
4.55
5.60
30s
19.44
16.76
16.79
40s
30.90
30.42
31.34
50s
26.08
28.81
26.12
60 and above
16.73
19.34
19.22
Mean (S.D.)
42.60 (60.27)
48.96 (11.58)
48.38 (12.73)
Household income
<2 million yen
6.3
5.04
3.73
2 million yen to <3 million yen
7.0
7.15
2.61
3 million yen to <4 million yen
9.41
10.74
6.53
4 million yen to <5 million yen
9.72
11.98
8.02
5 million yen to <6 million yen
9.43
12.35
11.94
6 million yen to <7 million yen
7.60
10.29
10.63
7 million yen to <8 million yen
7.22
9.99
9.70
8 million yen to <9 million yen
5.43
7.63
9.89
9 million yen to <10 million yen
5.27
7.54
9.70
≥10 million yen
11.77
17.28
27.24
Do not want to answer
20.86
-
-
Mean (S.D.)
472.69 (344.31)
720.13 (459.77)
963.25 (699.01)
Drive time
<15 min
9.67
11.29
9.70
15 min to <30 min
23.63
29.48
27.43
30 min to <1 h
26.22
34.01
31.16
1 h to <2 h
12.26
15.95
16.23
2 h to <3 h
3.94
4.99
7.84
3 h to <4 h
1.51
1.93
3.73
4 h to <5 h
0.66
0.83
1.31
5 h to <6 h
0.37
0.43
0.19
≥6 h
1.00
1.09
2.43
Do not have a car
20.75
-
-
Mean [min] (S.D.)
56.22 (60.27)
56.25 (58.56)
67.99 (72.00)
Number of owned cars
0
20.75
-
-
1
49.76
62.09
30.04
2
21.54
28.23
43.66
3
5.29
6.61
13.81

4
1.76
2.11
4.85
≥5
0.91
0.96
7.65
Mean (S.D.)
1.20 (0.95)
1.52 (0.79)
2.16 (1.13)
* N means the sample size.

Figure
Figure 1 SEM path diagram of Model I (N = 106, 446)
GFI = 0.915, AGFI = 0.894, CFI = 0.864, RMSEA = 0.055
Standard errors in parentheses: *** p < 0.01, ** p < 0.05, * p < 0.1
EE9
EE1
EE2
EE3
EE4
Evaluation
of EVs
EA7
EA8
EA1
EA2 EA3
Environmental
awareness
EA4 EA5 EA6
Purchase
intention
ε22
Age
Number of
Vehicles Gender
Drive time Household
income
ε4
ε21
ε20
.63
.7
.79
.85
.82
.65
.74
.67
.59
.26
.35
.33
.42
.01***
(.003)
-.02
(.003)
.02***
(.003)
.23***
(.003)
-.14***
(.003)
-.001
(.003)
.07***
(.003)
.09***
(.004)
.04***
(.006)
.06***
(.004)
-.008**
(.003) .09***
(.004)
.29***
(.004)
Actual use of Automobiles Individual attributes
ε8
ε3
ε2
ε1
ε5 ε6
ε7
EE12
EE11
EE10
EE8EE7
EE5 EE6
ε23
ε19
ε18
ε17
ε16
ε15
ε14
ε13
ε12
ε10
ε11
ε9
.56 .39 .17 .54
.52
.59
.3

Figure 2 SEM path diagram of Model II (N = 536)
GFI = 0.858, AGFI = 0.824, CFI = 0.867, RMSEA = 0.061
Standard errors in parentheses: *** p < 0.01, ** p < 0.05, * p < 0.1
EE9
EE1
EE2
EE3
EE4
Evaluation
of EVs
EA7
EA8
EA1
EA2 EA3
Environmental
awareness
EA4 EA5 EA6
Satisfacti on
ε22
Age
Number of
Vehicles Gender
Drive time Household
income
ε4
ε21
ε20
.69
.72
.89
.85
.82
.68
.8
.65
.52
.34
.57
.46
.57
-.06
(.05)
-.07
(.05)
-.02
(.05)
.15***
(.05)
-.08*
(.04)
-.05
(.05)
-.0007
(.05)
.066
(.05)
.07
(.1)
.2***
(.05)
-.12**
(.05)
.28***
(.05)
.29***
(.05)
Actual use of Automobiles Individual attributes
ε8
ε3
ε2
ε1
ε5 ε6
ε7
EE12
EE11
EE10
EE8EE7
EE5 EE6
ε23
ε19
ε18
ε17
ε16
ε15
ε14
ε13
ε12
ε10
ε11
ε9
.46 .61 .29
.57
.36
.47
.22

Appendix
Appendix 1 Environmental awareness of respondents (Model I, N = 106,446)
Environmental awareness
(Model I, N = 106,446) (%)
Very
important
Somewhat
important
Neither
Not very
important
Not at all
important
The level of importance for
each of the following items
about environmental policy
The percentage of renewable
energy generation amount to
total generated power amount
(EA1)
15.74
44.78
30.56
5.72
3.20
The percentage of eco cars in
total number of car owners
(EA2)
7.93
39.86
39.08
8.88
4.25
Annual emissions of GHGs
(EA3)
16.99
45.28
29.11
5.53
3.09
The percentage of
endangered species in
vertebrates (EA4)
9.58
34.35
44.11
8.39
3.58
The final disposal of garbage
(EA5)
17.44
50.66
26.60
3.53
1.76
The percentage of
reuse/recycling (EA6)
14.69
49.61
29.52
4.18
1.99
Pollution index of rivers and
lakes (BOD) (EA7)
17.71
48.78
28.52
3.47
1.52
The concentration of PM2.5
(EA8)
22.95
47.69
24.69
3.32
1.34

Appendix 2 Environmental awareness of respondents (ModelⅡ, N = 536)
Environmental awareness
(Model II, N = 536) (%)
Very
important
Somewhat
important
Neither
Not very
important
Not at all
important
The level of importance for
each of the following items
about environmental policy
The percentage of renewable
energy generation amount to
total generated power amount
(EA1)
23.51
44.03
23.88
5.22
3.36
The percentage of eco cars in
total number of car owners
(EA2)
16.98
42.54
29.10
6.72
4.66
Annual emissions of GHGs
(EA3)
22.20
41.60
26.68
6.16
3.36
The percentage of
endangered species in
vertebrates (EA4)
13.62
32.46
40.49
9.14
4.29
The final disposal of garbage
(EA5)
20.52
49.25
23.13
4.48
2.61
The percentage of
reuse/recycling (EA6)
17.35
49.63
25.75
3.92
3.36
Pollution index of rivers and
lakes (BOD) (EA7)
19.40
48.32
26.31
3.54
2.43
The concentration of PM2.5
(EA8)
25.75
45.15
25.19
2.24
1.68

Appendix 3 Evaluation of EVs
Evaluation of EVs (%)
Model I
(N = 106, 446)
Model II
(N = 536)
Benefits of purchasing EVs
(Cronbach’
alpha = 0.709)
(Cronbach’
alpha = 0.763)
Good fuel consumption (EE1)
56.30
55.41
Eco-friendly (EE2)
60.26
63.81
Less traveling sound and vibration (EE3)
30.72
52.24
Can be charged at home (EE4)
23.69
21.46
Can be charged in various place such as convenience store
and shopping centers (EE5)
38.66
58.40
Enjoyable ride (EE6)
15.48
26.87
Defines status (EE7)
5.67
31.34
Good acceleration performance (EE8)
1.13
6.53
Government subsidies can be obtained (EE9)
2.47
5.97
Can be used as an external power source (EE10)
7.98
38.81
Nighttime electric power can be used effectively (EE11)
15.71
32.46
The interior space is wide (EE12)
12.52
13.25
Do not know
17.55
30.22

Appendix 4 Purchase intention and post-purchase satisfaction of respondents
Purchase intention and the level of satisfaction of EVs
(%)
Model I
(N = 106,446)
Model II
(N = 536)
Your point of view on purchasing EVs
I am considering
5.54
I am not considering
94.46
Your point of view on the level of satisfaction of EVs
Completely satisfied
28.73
Slightly satisfied
44.40
Neither
12.50
Slightly dissatisfied
10.63
Completely dissatisfied
3.73
Mean (S.D.)
0.055 (0.229)
3.838 (1.072)

Appendix 5 Results of Model I (standardized solutions)
Variables
Coefficients
Standard
error
Z
P > |z|
95% Confidence
interval
Evaluation of EVs ←
environmental awareness
0.287
0.004
81.26
0.000
0.280
0.294
Evaluation of EVs ←
number of owned cars
0.071
0.003
20.83
0.000
0.064
0.078
Evaluation of EVs ← drive time
0.002
0.003
7.13
0.000
0.016
0.029
Evaluation of EVs ← age
0.063
0.004
17.23
0.000
0.055
0.070
Evaluation of EVs ← gender
0.093
0.004
26.02
0.000
0.086
0.100
Evaluation of EVs ←
household income
-0.008
0.003
-2.20
0.028
-0.014
-0.001
Environmental awareness ←
Number of owned cars
0.022
0.003
7.13
0.000
0.016
0.029
Environmental awareness ←
Drive time
0.013
0.003
4.02
0.000
0.006
0.019
Environmental awareness ← age
0.230
0.003
72.92
0.000
0.223
0.236
Environmental awareness ← gender
-0.137
0.003
-42.39
0.000
-0.144
-0.131
Environmental awareness ←
household income
-0.001
0.003
-0.31
0.760
-0.007
0.005
Purchase intention ←
evaluation of EVs
0.087
0.004
23.60
0.000
0.080
0.095
Purchase intention ←
environmental awareness
0.038
0.006
6.77
0.000
0.027
0.049
EE1 ←evaluation of EVs
0.264
0.003
79.61
0.000
0.258
0.271
EE2 ←evaluation of EVs
0.332
0.003
102.58
0.000
0.325
0.338
EE3 ←evaluation of EVs
0.424
0.003
140.48
0.000
0.418
0.430
EE4 ←evaluation of EVs
0.585
0.003
226.62
0.000
0.580
0.590
EE5 ←evaluation of EVs
0.558
0.003
210.19
0.000
0.552
0.563
EE6 ←evaluation of EVs
0.388
0.003
124.89
0.000
0.382
0.394
EE7 ←evaluation of EVs
0.166
0.003
48.40
0.000
0.159
0.173
EE8 ←evaluation of EVs
0.350
0.003
110.12
0.000
0.343
0.355
EE9 ←evaluation of EVs
0.538
0.003
199.17
0.000
0.533
0.543
EE10 ←evaluation of EVs
0.521
0.003
187.56
0.000
0.516
0.527
EE11 ←evaluation of EVs
0.587
0.003
226.27
0.000
0.582
0.592
EE12 ←evaluation of EVs
0.304
0.003
93.19
0.000
0.297
0.310
EA1 ← environmental awareness
0.671
0.002
361.80
0.000
0.668
0.675
EA2 ← environmental awareness
0.631
0.002
312.20
0.000
0.627
0.635
EA3 ← environmental awareness
0.739
0.002
470.69
0.000
0.736
0.742
EA4 ← environmental awareness
0.646
0.002
332.58
0.000
0.642
0.649
EA5 ← environmental awareness
0.821
0.001
689.42
0.000
0.819
0.823
EA6 ← environmental awareness
0.850
0.001
806.96
0.000
0.848
0.852
EA7 ← environmental awareness
0.794
0.001
600.88
0.000
0.791
0.796
EA8 ← environmental awareness
0.703
0.002
408.29
0.000
0.700
0.706

Appendix 6 Results of Model II (standardized solutions)
Variables
Coefficien
ts
Standard
Error
z
P > |z|
95% Confidence
Interval
Evaluation of EVs ←
environmental awareness
0.287
0.465
6.16
0.000
0.195
0.378
Evaluation of EVs ←
number of owned cars
-0.001
0.468
-0.02
0.987
-0.092
0.091
Evaluation of EVs ← drive time
-0.074
0.047
-1.58
0.115
-0.167
0.018
Evaluation of EVs ← age
0.202
0.047
4.32
0.000
0.110
0.293
Evaluation of EVs ← gender
0.066
0.046
1.43
0.152
-0.024
0.157
Evaluation of EVs ←
household income
-0.117
0.048
-0.97
0.334
-0.138
0.047
Environmental awareness ←
number of owned cars
-0.020
0.046
-0.45
0.650
-0.110
0.069
Environmental awareness ←
drive time
-0.058
0.046
-1.26
0.208
-0.148
0.032
Environmental awareness ← age
0.154
0.045
3.41
0.001
0.065
0.242
Environmental awareness ← gender
-0.080
0.045
-1.77
0.076
-0.168
0.008
Environmental awareness ←
household income
-0.045
0.047
-0.97
0.334
-0.138
0.047
Satisfaction ←
evaluation of EVs
0.279
0.049
5.71
0.000
0.183
0.375
Satisfaction ←
environmental awareness
0.071
0.100
0.71
0.480
-0.125
0.266
EE1 ←evaluation of EVs
00337
0.044
7.71
0.000
0..251
0.423
EE2 ←evaluation of EVs
0.464
0.041
11.43
0.000
0.385
0.544
EE3 ←evaluation of EVs
0.568
0.036
15.88
0.000
0.498
0.638
EE4 ←evaluation of EVs
0.521
0.038
13.75
0.000
0.446
0.595
EE5 ←evaluation of EVs
0.462
0.040
11.54
0.000
0.383
0.540
EE6 ←evaluation of EVs
0.611
0.034
18.09
0.000
0.545
0.677
EE7 ←evaluation of EVs
0.294
0.453
6.49
0.000
0.205
0.383
EE8 ←evaluation of EVs
0.573
0.035
16.26
0.000
0.504
0.643
EE9 ←evaluation of EVs
0.572
0.035
16.15
0.000
0.503
0.642
EE10 ←evaluation of EVs
0.361
0.043
8.34
0.000
0.276
0.446
EE11 ←evaluation of EVs
0.467
0.040
11.69
0.000
0.389
0.545
EE12 ←evaluation of EVs
0.220
0.047
4.72
0.000
0.129
0.312
EA1 ← environmental awareness
0.650
0.027
24.06
0.000
0.597
0.703
EA2 ← environmental awareness
0.688
0.025
27.51
0.000
0.639
0.737
EA3 ← environmental awareness
0.801
0.018
44.80
0.000
0.766
0.737
EA4 ← environmental awareness
0.680
0.025
26.87
0.000
0.630
0.729
EA5 ← environmental awareness
0.824
0.016
50.18
0.000
0.791
0.856
EA6 ← environmental awareness
0.849
0.015
57.77
0.000
0.820
0.878
EA7 ← environmental awareness
0.814
0.017
48.06
0.000
0.781
0.847
EA8 ← environmental awareness
0.716
0.023
30.89
0.000
0.670
0.761