Đề tài nghiên cứu factors affecting decisions to use green sm bike service of students in ho chi minh city

The proposed research model 21 2.7.1, Impact of Environmental Concern on Pro-environmental Behavior.... Health Consciousness Perceived Behavior Control Subjective Norms Pro-environmental

BỘ GIÁO DỤC VÀ ĐÀO TẠO DAI HOC KINH TE THANH PHO HO CHI MINH UEH

UNIVERSITY

`

Đề tài nghiên cứu FACTORS AFFECTING DECISIONS TO USE GREEN SM BIKE SERVICE OF STUDENTS IN HO CHI MINH CITY

Mon : Nghiên cứu Marketing Giảng viên hướng dẫn : Trần Mai Đông

Lớp : KM002

Mã lớp học phần : 24D1MAR50301708

Nhóm thực hiện :_ Nguyễn Thị Bích Trâm — 31221026739

Nguyễn Thi Ngoc Han — 31221022300

Tran Gia My — 31221021933

Chau Ngoc Lan — 31221026723 Dinh Thi Thuy Dung — 31221022296 Nguyễn Thị Kim Liên - 31221026400

Pham Ngan Giang — 31221023750

Thành phố Hô Chí Minh, ngày 15 tháng 4 năm 2024

Acknowledgements Our research is conducted from the Marketing Research subject taught by Mr Tran Mai Dong He helped and supported our group in this direction and topic He helped direct our interest in topics related to green consumption behavior, so that we had more ideas especially for electric vehicles Thank you very much to our professor, without your sincere comments and academic and emotional support, we would not have been able to complete this article!

We also want to extend our sincere thanks to our families and all of our special friends Thank you for always standing by, providing moral support and supporting us with all your efforts during the data survey process, thank you!

Table of contents

Acknowledgements 2 List of abbreviations 6 List of tables 7 List of figures 8 Abstract 10 CHAPTER 01: INTRODUCTION 11 1.1 Research background and statement of the pro bÌlem 55555555 sss II 1.2 Research objectives 12 1.3 Research objects 14 1.3.1 Research subject 14 P.27 Y1 - 14 1.4 Research method 15 1.41, Qualitative research method: " 15 1.4.2 Quantitative research methods 15 1.5 Research structure 15 CHAPTER 02: LITERATURE REVIEW AND HYPOTHESIS DEVELOPMENT

16 2.1 Foundation Theory 16 2.2 Electric Vehicles 17 2.3 Green SM Bike 17 2.4, Pro-environmental Behavior 18 2.5 Green consumption 19 2.6 Prior relevant studies 19 2.7 The proposed research model 21 2.7.1, Impact of Environmental Concern on Pro-environmental Behavior 21 2.7.2 Impact of Health consciousness on Pro-environmental Behavior 21 2.7.4 hmpact 0ƒ Subjective Norm on Pro-environmenfal Belqvior « 22

3

2.7.5 Impact of Perceived Behavioral Control on Pro-environmental Behavior

2.7.6 Impact of Pro-environmental Behavior on 2.8 Research model

2.9 Summary CHAPTER 03: RESEARCH METHODOLOGY,

3.2 Quantitative Methods 3.3 Data analysis process

3.3.1 Descriptive statistics analysis 3.3.2 Measurement Model

3.3.3, Assessing Conbach’s Alpha COGfÍTCÍCGHÍ HH HH ng mg se 3.3.4 Acessing EFA exploratory factor

3.3.5 Assessing Composite Reliability

3.3.6 Assessing Convergent Validity

3.3.7, Assessing Discriminant Validity

3.4 Assessing Structural model 3.4.1, Assessing Multicollinearity

3.4.2 Relationship in structural model

3.4.3 Assessing Coefficient of determination (R9

3.4.4, Assessing Effect Size (ff 3.4.5 Assessing the relevance of (@)

3.5 Measurement scale 3.6 Sample characteristics 3.7 Sample demographic characteristics 3.8 Summary

CHAPTER 04: DATA ANALYSIS AND RESULTS

4.1 Assessment of measurement scales

4.1.1, Cronbach's Alpha reliability test 40

Health Consciousness Perceived Behavior Control Subjective Norms Pro-environmental Behavior Green Consumption Heterotrait-Monotrait Ratio Partial least squares Structural equation modeling Variance inflation factor

Average Variance Extracted Exploratory Factor Analysis

Sample demographic characteristics Cronbach's Alpha reliability test EFA exploratory factor analysis results for independent variables EFA exploratory factor analysis results for other variables

CR composite reliability assessment AVE assessment

Fomell-Larcker Criterion Heterotrait-Monotrait Ratio (HTMT) Inner VIF assessment

Structural model quality Structural model results Results of testing the magnitude of the effect f in the model

List of figures Figure 1.1 The research framework and hypothesis

Figure 3.1 Research process

Figure 4.1 The impact of factors

Factors affecting decisions to use Green SM bike service of students

in Ho Chi Minh city

Abstract

In the modern context, the use of high-tech mobility services is becoming increasingly popular, especially among students - who are always looking for convenience and efficiency Green SM, the first new technology motorbike taxi service in Vietnam using electric vehicles, has attracted the attention of students in Ho Chi Minh City, especially the students of University of Economics Ho Chi Minh City (UEH) thanks to its commitment to choosing friendly means of transportation with the environment This research article aims to learn and analyze the factors that influence the decision to use Green SM Bike services of more than

200 UEH students, thereby providing a deeper insight into consumer behavior as well as the UEH students’ relationship with green consumption and environmental protection

Keywords: Green SM, green consumption, green technology transfer, electric vehicles, pro-

environmental behavior

CHAPTER 01: INTRODUCTION

1.1 Research background and statement of the problem

In recent years, the use of technological motorbike taxis (such as GrabBike, Gojek., ) has become an indispensable part of daily life to meet the travel needs of people in Vietnam The rapid development of information technology and smartphones has created an infrastructure for this service to develop strongly, while meeting people's transportation needs

in a convenient and flexible way

On the other hand, the living environment and environmental protection have become one of the top concerns of people around the world The topic of greenhouse gas (GHG) emissions is currently one of the most hotly contested topics on the planet It has been claimed that the most abundant greenhouse gas (GHG) in the atmosphere is carbon dioxide (CO2) The transportation industry is the primary source of these pollutants (World Health Organization [WHO], 2019) Since they can replace gasoline-powered vehicles with electric vehicles (EVs), which can reduce GHG emissions (Bhutto et al., 2020), EVs can be seen as technological solutions to the problem of GHG emission (Asamer et al., 2016) Green transportation is one of the effective solutions to reduce greenhouse gas emissions, contributing to responding to climate change, protecting the environment and human health According to the U.S Department of Energy, green transportation can reduce carbon dioxide emissions by 50% and nitric oxide emissions by 90% compared to gasoline-powered transportation Green transportation is also one of the areas of concern and priority at the COP

26 Climate Change Summit taking place in November 2023 in Glasgow, UK At this conference, countries committed to cooperate to transition from coal-based power generation

to clean energy in the 2030s for major economies and in the 2040s globally Vietnam is one of the countries that have signed this Declaration, and committed to achieving net zero emissions

by 2050

However, in Vietnam, air pollution caused by motor vehicles is increasingly serious Many studies have shown that air pollution in major cities in Vietnam exceeds permissible levels, the main cause of air pollution from traffic activities, accounting for 60-70% According to the Ministry of Natural Resources and Environment, Vietnam emitted about 240 million tons of carbon dioxide into the air in 2020, of which the transportation sector accounted for about 12% Therefore, concern for the environment has made a part of motorbike taxi users focus on minimizing negative impacts on the environment By using electric vehicles instead of gasoline and diesel vehicles The purpose of technological

10

developments, such as electric vehicles, is to solve emerging social and transportation problems (Oelschlaeger, M.A.X 2019), as well as reduce pollution levels (Ajanovic, A.; Haas,

a test of a delivery service using electric motorbikes from startup Selex Motors (Viet Hung, 2023) In addition, there are still other challenges such as the relatively high cost of electric vehicles and problems related to charging stations that can also disrupt transportation and delivery businesses, which are special factors important for the industry (Viet Hung, 2023) Xanh SM's strategic orientation is to become a leading company in the field of green and smart transportation in Vietnam, as well as Southeast Asia To achieve this goal, the company needs to obtain quality and updated scientific research on customer behavior, attitudes and factors that influence customers' decisions to use electric taxi services

What factors make consumers choose and trust Green SM Bike service? Up to now, there are many research articles on the behavior of using technology motorbike taxis, however, research articles on electric technology motorbike taxis are still few and have not clarified the factors influence usage decisions From the above reasons, the author conducted the topic: "Research on factors affecting the decision to use Green SM Bike service of students in Ho Chi Minh City" This research paper aims to explore and analyze the factors that influence students’ especially UEH students' decisions to use Green SM services, thereby providing insight into their green consumption behavior, as well as propose specific solutions

to promote the use of environmentally friendly products and services among the student community

11

1.2 Research objectives

Research on factors affecting UEH students' decisions to use Green SM Bike service plays an important role in correctly identifying consumer behavior and supporting administrators in decision making Marketing strategies, this research process contributes to practice as well as science:

In terms of practice, this research provides insight into how young people in Ho Chi Minh City access and use SM Bike green services, understanding the attitudes, behaviors, desires and needs of young people Ho Chi Minh City youth for SM Bike Green service Including identifying factors that have a strong impact on service usage decisions such as health concern, environmental concern, attitude, subjective norms, percetved behavioral control and perceived consumer effectiveness, thereby supporting administrators to come up with marketing strategies and messages they want to convey to improve service quality and meet customers' needs and desires for SM Green services GSM's Bike, creating a better experience for service users Compared to traditional motorbike taxi services or popular technology vehicle services today that use gasoline and oil as fuel, increasing the level of dust and air pollution, Xanh SM Bike's service wants to bring many benefits benefits the community, contributes to reducing air and noise pollution, while saving energy and reducing costs for users, promoting sustainable transportation development, contributing to implementing Vietnam's commitment to Reduce greenhouse gas emissions, contributing to reducing fossil resource exploitation Aiming to connect and form a green and smart mobility ecosystem across Vietnam in the future

Another thing the study shows is that the GSM brand with the SM Bike Green service has partly supported the government in implementing state regulations and recommendations

on air environmental protection and anti-pollution greenhouse gas, helping to save national fossil resources and develop the country's economy to become more modern and green About science: research helps contribute knowledge to the fields of marketing and science Additional factors affecting the decision to use electric motorbike services for young people in Ho Chi Minh City The results of this study can provide insights into how young people evaluate and prioritize these factors in the process of choosing and using electric motorbike services Opens up new theoretical development directions on young consumer behavior, provides a model and decision-making process for using Green SM Bike service, thereby serving as a basis for further research or supporting consultation Reference for related marketing research: purchase intention, intention to use other services in the

12

industry, such as intention to use natural products such as paper straws, paper cups in food service , or in the tourism industry Opens up new discoveries in consumer behavior in different research groups according to age, occupation, education level, etc In addition, it also contributes to green consumption behavior, protecting the environment

Conclusion: With the benefits that this research article brings, the research team wants

to send out the message that: Xanh SM Bike's services carry the mission of creating a green, clean, and beautiful society for the community, helping to protect the community environment of emissions from vehicles using fossil fuels That's why the research team hopes that the government will propose policies to encourage people to use green, environmentally friendly services suitable for social development

of economic sectors has attracted a large number of young people to study and work Therefore, this is a suitable area to conduct in-depth research on the factors that influence young people's usage decisions At the same time, because the authors of this research article are studying at UEH University of Economics Ho Chi Minh City, it will be easy to collect survey data from 250 UEH students This number not only ensures representativeness of the student population but also allows for detailed and accurate data analysis, thereby drawing valuable conclusions

13

About time: Data was collected over 3 days in 2024 (March 21st, 2024 to March 23rd, 2024) at different times of days The studies and published results of the author’s group were carried out from March to April 2024

1.4 Research method

1.4.1 Qualitative research methods

We used qualitative research methods to understand and identify factors that influence the decision to use Xanh SM services In the first stage, the measurement scales applied to all research constructs were adopted from previous studies, translated into Vietnamese and adjusted to suit the subjects and research objectives Questions are rated on a Likert scale from | to 5, with 1 being completely disagree and 5 being completely agree Then our group discussed and designed the survey questionnaire

1.4.2 Quantitative research methods

The second phase is quantitative research including the main survey and data analysis The survey was designed on Google Form and sent to more than 250 UEH students through social networking platforms such as Facebook, Zalo, Instagram, We also conducted in-depth interviews with 10 UEH students using Green SM service, to supplement and clarify the results from quantitative research methods The interview questions were designed to explore students’ reasons, feelings, opinions and experiences when using the service Interviews were recorded and transcribed, then coded and analyzed using content analysis After collecting, the data set was analyzed using SPSS software including the following steps: scale, test for common method bias, and evaluate the structural model by hypothesis testing

Chapter 04 - Data Analysis and Results: This section analyzes the study's data set It includes the following steps: evaluating the scale, checking for common method bias, evaluating the structural model

14

Chapter 05 - Discussion and Conclusion: This final chapter summarizes the important results of the current thesis In addition, limitations of the study and suggestions for further research are also mentioned

CHAPTER 02: LITERATURE REVIEW AND HYPOTHESIS

DEVELOPMENT

In this chapter, we clarify concepts and theories related to the research topic based on previous research and make reasonable adjustments From there, develop hypotheses about the relationship between argument factors to decide to use Green SM Bike service of UEH students

2.1 Foundation Theory

The theory of planned behavior (TPB), developed from the theory of reasoned action (Ajzen, I 2020), explains environmental behavior based on attitudes and subjective standards, mediated by behavioral intentions with apparent behavioral control Theory of Planned Behavior Ajzens’- TPB and its variants have been the most commonly applied to green consumption (Biswas & Roy, 2015; Hanss et al., 2016; Peattie, 2010) The TPB successfully explained environmental behavior, including travel mode choice (Ajzen, I 2020), household, and general pro-environmental behavior (Yuriev, A et al., 2020) Using the TPB, researchers have examined alternative transportation, waste recycling, water conservation, energy conservation, low carbon consumption, and other pro-environmental behaviors (Effendi, M.I

et al., 2020) For electric vehicle adoption, several studies have demonstrated the relevance of single TPB predictors or the TPB as a whole (Moons & De Pelsmacker, 2012; Schmalfub et al., 2017; Wang, Fan, Zhao, Yang, & Fu, 2016) There has been an increase in the number of studies using the TPB to examine pro-environmental practices (Yuriev, A et al., 2020) Thus,

it is essential to have a greater understanding of the factors that impact the pro-environmental behaviors (PEBs) of individuals, particularly university students (Yusliza, M et al., 2020; Foster, B et al., 2022; Huang, J et al., 2020) The positive or negative evaluation of an activity is the attitude towards that conduct When people think highly of the results of their actions, they are more likely to continue taking those actions (Ajzen, I 2020) A person’s likelihood of engaging in a behavior changes (or does not change) depending on whether or not he or she feels that the action is acceptable to other people (Ajzen, I 2020) It's influenced

by one’s peers, family members, friends, or prominent members of the community, and exert pressure on people (Fishbein & Ajzen, 1975) Additionally, the addition of perceived

15

behavioral control extends the theory of reasoned action into the theory of planned behavior, whose predictive power has been significantly increased Perceived behavioral control belongs to the “rational-choice model,” which assumes that “people behave rationally and logically during the process of decision making” (Ajzen, 1991, p 182) Several management scholars (e.g Bhutto et al., 2020; Channa et al., 2020; Kumar et al., 2017) have frequently used the theory of planned behavior and have found that perceived behavioral control 1s its fundamental factor There is a consensus in the consumer behavior-focused psychological literature that the TPB’s predictive ability is enhanced when complemented by other variables (Karimi, S et al., 2022) All these other concepts have to do with the environment while discussing sustainability (Yafi, E et al., 2021)

2.2 Electric Vehicles

An electric vehicle (EV) is a type of vehicle that runs solely on electricity, without relying on any other source of power such as gasoline or diesel These vehicles utilize electric motors powered by rechargeable battery packs, allowing them to operate without emitting tailpipe pollutants

EVs are propelled by electric motors, and power is maintained by a portable energy storage device or rechargeable battery These cars utilize less energy, produce less greenhouse gas emissions, and produce less noise The various EV classifications are as follows:

HEV: Hybrid electric vehicles (HEVs) combine the power of an electric motor and an engine to run on both gasoline and electricity The battery is charged by the energy produced

by the braking system

PHEV: Plug-in hybrid electric vehicles, or PHEVs, are similar to HEVs but feature bigger batteries and a smaller engine Either the braking system or an external electric charging outlet can be used to recharge the batteries

BEVs: These vehicles don't have an engine and move forward using electric motors and batteries for energy storage They rely on outside power outlets to replenish the battery These automobiles are often referred to as battery electric vehicles (BEVs), EVs, or plug-in vehicles 2.3 Green SM Bike

Green SM Bike, Vietnam’s first pure electric motorbike taxi service, was officially launched on August 14, 2023 by the Green and Smart Mobility Joint Stock Company (GSM), following the roll-out of its e-taxi hailing service last April this year

16

Green SM's adoption of the electric motorcycle as a transportation solution is a significant step toward the government's objective of achieving net zero (Net Zero) emissions

by zero At the same time, Green SM pledges to donate 2050 dong to the Fund for a Green Future in order to support environmental projects for every journey that customers choose to take on a Green SM Bike

The goal of the Green SM Bike service, together with Green SM Taxi and Green SM Luxury, is to reimagine public transportation in order to promote green living and the use of electric vehicles in the community This is a generation of ecologically friendly, health- conscious, and noiseless-running cars that don't smell like gasoline

2.4 Pro-environmental Behavior

According to Stern (2000) and Kollmuss and Agyeman (2002), deliberate action that can lessen a negative influence on the environment is the generally accepted definition of pro- environmental behavior A variety of operationalized behaviors are included in the category of pro-environmental behavior, including recycling (Hansmann et al., 2006; Kléckner and Oppedal, 2011; Byme and OdRegan, 2014; Zhang et al., 2016; Fu et al., 2017), using transportation (Eriksson et al., 2008), waste management (Begum et al., 2009; Rigamonti et al., 2014; Sasaki and Araki, 2014; Lobato et al., 2015; Pdéldnurk, 2015; Liu et al., 2017), energy consumption (Tester, 1992; Berardi, 2017), buying environmentally friendly products (Ramayah et al., 2010), and electrical appliances (Shih, 2001; Aizawa et al., 2008) No one seems to have a significant disagreement or controversy regarding the definition of pro- environmental behavior It focuses on the autonomy of actors and decreases the damage to the world Additionally, based on the perspective of the impact of behavior on the environment, the definition of pro-environmental behavior is then extended to minimize the damage to the environment and even gain from it (Steg et al., 2009) This definition centers around enhancing environmental conditions while also attempting to minimize the negative effects of the environment, including greenhouse gas emissions, wasted natural resources, and so on Furthermore, from a sustainability perspective, pro-environmental behavior is behavior that promotes environmental sustainability (Mesmer-Magnus et al., 2012) On the whole, this paper asserts that pro-environmental behavior encompasses actions taken with a conscious intention to protect the environment and enhance its sustainability

Based on the definitions above, there are several labels that are similar to environmentally responsible behavior in the literature, these include "ecological behavior",

"environmental behavior", "environmental action", "responsible environmental behavior",

17

"ecological behavior", "green behavior", and "sustainable behavior" The variety of different labels associated with construction is primarily accidental, and there is a shared correspondence in several areas (Dilchert et al., 2012) For instance, they share the same specific behaviors like reducing the amount of resources used

2.5 Green consumption

Green consumption is defined as consumer behavior that reduces the harm that products cause to the environment throughout their purchase, use, and disposal (Peattie, 2010) According to ElHaffaret al (2020) and Liobikiené & Bernatoniené (2017), it helps strike a balance between addressing personal requirements and decreasing environmental issues by limiting the environmental impact of consumption without compromising quality of life The term “green consumption” may not always have a clear and consistent meaning (Peattie, 2010) In fact, it has been used differently and sometimes interchangeably with other terms (Kim et al., 2012) including socially responsible consumption (Antil, 1984), ecologically conscious (Fraj & Martinez, 2006), environmentally responsible consumption (Gupta & Ogden, 2009), environmentally friendly consumption (Laroche et al., 2001) and environmentally friendly consumption (Welsch & Kiihling, 2009) Furthermore, the term green consumption itself can be problematic because while “green” implies conservation of environmental resources, “consumption” involves some form of destruction (Peattie, 2010) Fortunately, there is a common theme running through all the terms: the desired consumption goal to minimize environmental consequences (Kim et al., 2012) At that time, the term

"green" can also be understood more broadly as "sustainable development orientation" (Peattie, 2010, p 197) Thus, in this article, green consumption is understood as an individual's consumption behavior related to environmental and resource issues and is motivated not only by the desire to satisfy the individual's needs but also a concern for the welfare of society as a whole (Antil, 1984; Antil & Bennett, 1979) This definition is similar

to the definition of sustainable consumption published by the United Nations National Environment Program in 1994 (see Ceglia et al., 2015; Liu et al., 2017) The development of smart green technologies, with a low carbon footprint, is especially important to meet future challenges towards a sustainable society (Biresselioglu et al., 2018; Carlucci, F., 2018) Reducing CO2 emissions 1s a global problem that can be partly solved by replacing fossil fuel vehicles with electric vehicles Therefore, using electric vehicles is considered a green consumption action in this study

18

2.6 Prior relevant studies

Table 2.1 Relevant empirical studies green consumption

Reference Independent Mediator/ Dependent Key finding

variables Moderator variables

Social and personal Pro- identities, as well as Gadenne, D et Subjective N/A environmental | values, can be reflected

al (2011) norm behavior in an individual's

consumption patterns Who are concerned Pro- about the environment Lee, Y etal | Environmenta N/A environmental | are more likely to

(2014) l concern behavior engage In pro-

environmental behaviors

Green consumption refers to consumer behavior that Pinto, C., D et Pro- N/A Green minimizes the negative

al (2014) environmental Consumption | environmental impact

Behavior of purchasing, using,

and disposing of products

There is a line of evidence about the Shimoda, A et Health Pro- positive associations

al (2020) consciousness N/A environmental | between health

behavior consciousness and pro-

environmental behaviors

Perceived behavioral Perceived Pro- control (PBC) relates Effendi, M., I behavioral N/A environmental | to intentions as well as

et al (2020) control behavior directly to pro-

environmental behavior

2.7 The proposed research model

2.7.1 Impact of Environmental Concern on Pro-environmental Behavior

The present study employs a conceptualization of environmental concern, drawing on previous research (Chan, 2001; Ishaswini & Saroj, 2011; Mainieri, Barnett, Valdero, Unipan,

& Oskamp, 1997; Pinto et al., 2011), to represent consumers’ overall attitude toward the environment and their level of concern about environmental threats Several research that looked at the role of environmental concern on pro-environmental behavior (Chan, 2001; Mainieri et al., 1997; Pinto et al., 2011) discovered that it had a substantial impact Ellen et al (1991), for example, found that environmental concern significantly influences a variety of environmental behavior (e.g., green buying, recycling, writing public authorities) Ellen et al (1991) separated the construct of environmental worry from PCE Similarly, there is a positive correlation between environmental concern and green purchasing behavior in India, according

to Ishaswini and Saroj's (2011) study According to these earlier research, people who care about the environment are more likely to adopt pro-environmental practices like using green services and buying green items As a result, we offer the following hypothesis:

Al: Environmental Concern positively impacts on Pro-environmental Behavior

2.7.2 Impact of Health consciousness on Pro-environmental Behavior

Health consciousness refers to a psychological state where an individual is aware of and involved in his/her health condition (Gould, 1990) There is a line of evidence about the positive associations between health consciousness and health behaviors (such as health information seeking, exercise and purchase of organic food) (Gould, 1990; Lockie et al., 2004; Iversen and Kraft, 2006); between health consciousness and pro-environmental behaviors (Ture And Ganesh, 2012); and health consciousness and anticonsumption attitudes

20

(.e an attitude that is opposed to consumerism, and the continual buying and consuming of material possessions) (Kaynak and Eksi, 2014) A previous study indicated that health- conscious customers are likelier to perform environmentally beneficial behavior than others (Rana and Paul, 2017), due to their information seeking behavior will be concerned of state of environment and its adverse impact on the individual or society in a large This may shape their belief about nature and human relationship, and may prompt them to take pro- environmental actions

2: Health Consciousness positively impacts on Pro-environmental Behavior

2.7.4, Impact of Subjective Norm on _ Pro-environmental Behavior

Subjective norm is associated with the emergence of social pressure while performing a certain behavior (Ajzen, 1991) Support and encouragement from family, close friends, coworkers, and media propaganda as attributes of subjective norms (Huang, X., Ge, J, 2019) contributed positively to the adoption of a technology (Yuen, K F et al., 2020) From the social-impact perspective, the individuals in a segment seem to be more closely connected to the other segment members than to non-segment members, and to be generally influenced by the opinions of the segment and by the normative pressure exerted by it (Ajzen, 1991, 2002) Social and community influences play a crucial role in shaping environmental behavior The intention to engage in certain actions is heavily influenced by social norms, as noted by Bamberg (2003) Adhering to behaviors considered norms within a group fosters a sense of belonging and group membership among individuals Social environments significantly impact the formation of people's values, with social influence exerting a positive effect on the intention to adopt environmental behaviors Strong social norms are often necessary to promote the adoption of various pro-environmental behaviors, as found by Ozaki (2011) Emotional, societal, and cultural factors also play a role in influencing consumers’ behavior regarding domestic energy use, as highlighted by Faiers et al (2007) Many individuals prioritize presenting a positive self-image, which motivates them to conform to the norms of their desired social group Social and personal identities, as well as values, can be reflected in

an individual's consumption patterns For instance, consumers perceive adopting green electricity as aligning with their pro-environmentalist identity (Ozaki, 2011) Research suggests that consumers are positively influenced by the opinions and actions of their family, friends, and associates (Jager, 2006; Pickett-Baker and Ozaki, 2008; Sidiras and Koukios, 2004), as well as by cultural values (Chan, 2001)

21

H3: Subjectives Norm positively impacts on Pro-environmental Behavior

2.7.5 Impact of Perceived Behavioral Control on Pro- environmental Behavior

Perceived Behavioral control shows a degree in which an individual feels that the appearance or failure of behavior in question is under his control (Ajzen, 1991) People tend not to form a strong intention to display a certain behavior if he believes that he does not have the source or opportunity to do so even though he has a positive attitude and he believes that other people who are important to him will approve of it PBC can influence behavior directly

or indirectly through intention The direct path from PBC to behavior is expected to occur when there is harmony between perceptions about control and the actual control of a person over a behavior Perceived behavioral control (PBC) relates to intentions as well as directly to pro-environmental behavior (Kaiser and Gutscher, 2003; Morren and Grinstein, 2016) Generally, PBC is measured by asking respondents how complicated specific or general behavior is performing Lakhan (2017) and Taylor and Todd (1995) studied communities with sophisticated waste management and found that PBC positively influences recycling behavior and intentions

4: Perceived Behavioral Control positively impacts on Pro-environmental Behavior 2.7.6 Impact of Pro-environmental Behavior on Green Consumption

Recently, society has become increasingly concerned about environmental protection (e.g Corraliza and Berenguer, 2000) As a result, many consumers are changing their consumption habits, choosing products that have less impact on the environment (Schaefer and Crane, 2005) Pro-environmental actions are actions that “intentionally aim to limit the adverse impact of one's actions on the natural and built world” (Ramkissoon, H 2020) It is consistent with the concept of "green consumption" Green consumption refers to consumer behavior that minimizes the negative environmental impact of purchasing, using, and disposing of products (Peattie, 2010) It minimizes the environmental impact of consumption without reducing quality of life (ElHaffar et al., 2020; Liobikiené & Bernatoniené, 2017), facilitating a balance between meeting personal needs and minimizing environmental problems Pro-environmental attitudes should be considered because they can lead to action (Balunde, A et al., 2020) Indeed, previous studies have revealed a statistical correlation between environmentally friendly ideas and ethical actions (El Batni, B 2019) The literature

22

on students' pro-environmental behavior is diverse In one line of research, students have been shown to care deeply about environmental issues (e.g., Michel, 2020)

5: Pro-environmental Behavior positively impacts on Green Consumption

2.8 Research model

Figure 1.1 The research framework and hypothesis

— Direct Environmental

H1 Health tomer 2 Consciousness rr “A Pro H6 Green

Environmetal Consumption

Behavior

H3 Subjective

Norms H4

Perceived Behavioral Control

Al: Environmental Concern positively impacts on Pro-environmental Behavior

H2: Health Consciousness positively impacts on Pro-environmental Behavior

H3: Subjective Norm positively impacts on Pro-environmental Behavior

4: Perceived Behavioral Control positively impacts on Pro-environmental Behavior 5: Pro-environmental Behavior positively impacts on Green Consumption

2.9 Summary

Overall, this chapter has presented research built on the foundation theory of TPB and evaluated each structure of the model As well as researches related to this project, 5 hypotheses are proposed and models related to factors affecting pro-environmental behavior

as well as green consumption were designed by our group The next chapter will be concerned with the methodology used for the present study

23

CHAPTER 03: RESEARCH METHODOLOGY

3.1 Procedure

Our research process includes the following steps:

The first step in this process was to review the literature and prior relevant papers (Table 2.1) to adopt the literature review of all the important definitions such as Pro-environmental Behavior, Electric Vehicles, Green SM Bike, Green Consumption and factors affecting UEH students’ decision to use SM green motorbike tax1

Second, build a research hypothesis: We build a research hypothesis based on theory and previous research to determine the relationship between variables such as Environmental Concern, Health Consciousness, Subjective Norms, Perceived Behavioral Control for Pro- environmental Behavior, Green Consumption

Third, construct the research model: Based on the researched hypothetical developments, we proceed to build a research model showing the impact of factors on pro- environmental behavior as well as green consumption In which factors such as Environmental Concern, Health Consciousness, Subjective Norm, Perceived Behavioral Control have a positive impact on Pro-environmental Behavior

Fourth, measurement scale & questionnaire: We adopt the measurement scale for all studied constructs from previous related studies; with some minor modifications to fit the

24

current research context Constructing an initial questionnaire in English from the original articles, then translated it into Vietnamese and then conducted a preliminary survey to adjust the questionnaire to ensure the logical and understandable meaning of each item and back to being translated into English for data analysis

Next, data collection: We collect data by sending an online survey to 250 UEH students who have used the Green SM Bike service at least once within 3 months The survey included questions evaluating research variables on a Likert scale from 1 to 5 In addition, we also conducted in-depth interviews with 10 students to collect more qualitative information

Then, analyze the data: We enter the collected data into SPSS 18.0 and SmartPLS 3.2.9 software to analyze using the analytical analysis method This method helps us test the influence of each independent variable on the dependent variable, as well as test the research hypotheses We also coded and analyzed the data using content analysis methods

Finally, conclusions and recommendations are given: We have summarized the main findings from the data analysis, as well as proposed solutions to improve services and provide sustainable consumption behavior

25

1 Literature Review

2 Build a research

hypothesis

3 Construct the research model

Questionnaire

7 Give the conclusions and

- Reliability (cronbach’s alpha and composite reliability)

- Convergent validity (AVE)

- Discriminant validity (cross loadings, Fornell-Larcker criterion, and the Heterotrait-Monotrait Ratio - HTMT)

- The collinearity issues (VIF value)

- The predictive power (R2) and predictive relevance

- Hypotheses testing (bootstrapping 1,000): direct effects, mediating effects, and the moderating effects

is 50, preferably 100 or more, and at least five observed variables for each measured variable (Hair et al., 2011) There are 22 observed variables in the research model that is presented in Chapter 2 Therefore, the minimum sample size should be: 5*22 = 110 The authors decided

to use a sample size of 200 persons in order to increase the reliability The author team will share the Google form link with friends and acquaintances over Zalo, Messenger, and Facebook There were 250 votes given in all, of which 200 were collected This survey was conducted between January 12, 2024, and March 23, 2024

This study uses the PLS-SEM method as the primary tool for data analysis for several reasons First of all, PLS-SEM is the preferred method because in a direct comparison with CB-SEM, the variance explained in the dependent variables is substantially higher (Hair et al., 2017), this method is more suitable when the purposes of the researchers are focusing on the predictive power of the dependent variable (Henseler et al., 2009) Furthermore, PLS-SEM has a relative advantage (compared to CB-SEM) in that it does not require the dataset to be normally distributed or has no multicollinearity problem (Hair et al., 2018) In addition, PLS- SEM can analyze models with many latent variables measured by many different parameters

at the same time, especially those measured by higher-order variables (Hair et al., 2017) Additionally, PLS-SEM allows both the measurement model and the structural model to estimate at the same time, avoiding skewed or inappropriate parts for the estimate (Hair et al.,

2018)

With the collected data set, after filtering and cleaning the data, the data will be processed using IBM SPSS 18.0 and SmartPLS 3.2.9 software SmartPLS 3.2.9 is used to analyze the accuracy of the measurement scales, R2 and f2.values The bootstrapping method was performed to test the significance of the path coefficients

3.3 Data analysis process

3.3.1 Descriptive statistics analysis

Summarize using a form taken from Google Form and perform testing and screening with Excel to avoid errors before conducting data analysis The final result is 200 samples We used SPSS 18.0 software to conduct descriptive statistical analysis to characterize the study sample Next, evaluate the research models through two types of models: (1) Effective index (Reflective measurement model) and (2) Composite index (Formal measurement model) due

to (Henseler & Chin, 2010) suggested

27

3.3.2 Measurement Model

The measurement model is evaluated based on reliability and validity In which, reliability is evaluated based on specific measures, Cronbach's Alpha reliability and Composite reliability coefficient (CR), and validity (convergent validity and discriminant validity) evaluated through Cross loading coefficient, Average Variance Extracted AVE and Correlation matrix between research variables

3.3.3 Assessing Conbach’s Alpha coefficient

The reliability of the scale is evaluated through the Cronbach's Alpha coefficient to evaluate the reliability of the scale based on internal consistency, that is, whether the observed variables in the scale are truly correlated or whether there is a close connection with each other or not (Nguyen Dinh Tho, 2014) The larger the Cronbach's Alpha coefficient, the higher the internal consistency reliability In addition to using Cronbach's Alpha coefficient, the reliability of the scale is also evaluated through the composite reliability coefficient According to researchers, composite reliability is a better evaluation index than Cronbach's Alpha because it does not make the mistake of assuming equal reliability of variables (Gerbing and Anderson, 1988) Composite reliability 1s best when it is greater than 0.7 (Hair

et al., 2010), however in exploratory research composite reliability can range from 0.6 to 0.7 (Hair et al., 2017) According to Hair et al (2017), when evaluating the internal consistency

of a scale, both Cronbach's Alpha coefficient and composite reliability criteria should be considered In CB-SEM, the composite reliability coefficient 1s calculated according to the formula of Joreskog (1971) based on the standardized regression coefficients of observed variables With PLS-SEM, the composite reliability coefficient is calculated according to the formula of Fornell and Larcker (1981) based on the different outer loadings of the latent variables

Cronbach’s Alpha coefficient takes values between 0 and 1 In general, an Alpha coefficient above 0.60 will pass the reliability test with values closer to | being preferable If this index is below 0.3, the variable will be eliminated from the study

3.3.4, Acessing EFA exploratory factor

By checking the reliability of the Cronbach Alpha scale, we are evaluating the relationship between variables in the same group and the same factor, not considering the relationship between all observations in the factors other Meanwhile, EFA considers the relationship between variables in all different groups (factors) to detect observed variables

28

that load on many factors or observed variables that are analyzed by the wrong factors from the beginning

Criteria in EFA analysis

The KMO coefficient (Kaiser-Meyer-Olkin) is an index used to consider the appropriateness of factor analysis The value of KMO must reach a value of 0.5 or higher (0.5

<KMO <1), which is a sufficient condition for factor analysis to be appropriate If this value 1s less than 0.5, factor analysis is likely not appropriate for the research data set

Bartlett's test of sphericity is used to see whether the observed variables in the factor are correlated with each other or not We need to note that the necessary condition to apply factor analysis is that observed variables reflecting different aspects of the same factor must be correlated with each other This point is related to the convergent validity in the EFA analysis mentioned above Therefore, if the test shows no statistical significance, factor analysis should not be applied to the variables under consideration The Bartlett test is statistically significant (sig Bartlett's Test < 0.05), proving that the observed variables are correlated with each other in the factor

Eigenvalue is a commonly used criterion to determine the number of factors in EFA analysis With this criterion, only factors with Eigenvalue > | will be retained in the analytical model

Total Variance Explained = 50% shows that the EFA model is appropriate Considering the variation as 100%, this value represents how much of the extracted factors are condensed and how much of the observed variables are lost

Factor Loading, also known as factor weight, this value represents the correlation between the observed variable and the factor The higher the factor loading coefficient, the greater the correlation between that observed variable and the factor and vice versa According to Hair et al (2010), Multivariate Data Analysis loading factor from 0.5 is a good quality observed variable, the minimum should be 0.3

® Factor Loading at + 0.3: Minimum condition for observed variables to be retained

e Factor Loading at + 0.5: Observed variable has good statistical significance

e Factor Loading at + 0.7: Observed variable has very good statistical significance

29

3.3.5 Assessing Composite Reliability

Composite reliability is a coefficient used to measure the internal consistency of indicators in a scale and is used as an alternative to Cronbach's Alpha coefficient (Netemeyer

et al., 2003)

The composite reliability coefficient CR uses standardized factor loadings and error variation of observed variables belonging to a latent variable McDonald (1970) gives the formula to calculate the composite reliability coefficient of a latent variable A including m observed variables as follows:

(Id, +Id, + +/d,,)”

~ (Id, +d, + +1d, +02 +0,°+ 40,7 m

CR

In there:

e CR: composite reliability CR of latent variable A

© Idi, Ido, Id: standardized loading factor of observed variable of latent variable A

® m-: number of observed variables of latent variable A

© 6/7, 62’, Gm’ | measurement error variance of observed variable belonging to latent variable A (Gn? = 1 —Idm’)

Like Cronbach’s Alpha, composite reliability values range from 0-1, with higher values closer to 1 indicating higher levels of reliability In particular, with exploratory research, a reliability value of 0.6 - 0.7 is accepted, however, while with many other studies, this value is required to be in the range of 0.7 - 0.9 to be effective accepted (Nunally and Bernstein, 1994),

If this value is greater than 0.95, it is considered problematic because there is a high possibility of overlapping observed variables, meaning the variables observe the same content If the composite reliability has a value less than 0.6, this indicates that there is a lack

of internal consistency reliability and should be reconsidered (Hair et al., A Primer on Partial Least Squares Structural Equation Modeling, 2014)

3.3.6 Assessing Convergent Validity

According to Hair and colleagues (2010), the standardized factor loading coefficient of the observed variables at the convergence threshold is greater than or equal to 0.50, Anderson and Gerbring (1988) said that the scale has convergent validity converge when the important standardized numbers of the scale are all greater than 0.50 and statistically significant (p- value < 0.05)

30

The average variance extracted AVE (Average Variance Extracted, symbol: pvc) or average variance extracted is used to check the convergence of observations within the same concept Variance extracted AVE was calculated for each unidimensional factor According to Bagozzi and Yi (1988), the average variance extracted AVE needs to be greater than 0.50 to satisfy convergent validity, Fornell and Larcker (1981) said that the variance extracted must

be greater than or equal to 0.50 to satisfy the requirement demand, while greater than 0.70 is considered good Just like the composite confidence level, this method is completely based on

a system of standardized regression numbers (Standardized Regression Weights)

Formula to calculate quoted variance value:

In there:

© px 1s the average variance extracted;

® 71s the observed variable;

e i, 1s the standardized regression coefficient in the linear structural model of the ith observed variable;

e 1-4/ is the measurement error variance of the ith observed variable

3.3.7 Assessing Discriminant Validity

Discriminant validity is the degree to which a concept is truly distinct from another concept by empirical standards (Hair et al., 2010) With CBSEM, the discriminant value of the scale is evaluated based on the criteria that the correlation between two concepts must be less than 0.85 (Hair et al., 2010; Kline, 2011) and the square of the correlation coefficient (maximum shared variance - MSV) is smaller than the AVE index In addition, another way to evaluate is to use the Fornell-Larcker criterion, according to which the Square root of AVE coefficient must be greater than the inter-construct correlations (Fornell and Lacker, 1981) With the PLS-SEM method, in addition to using the Fomell-Larcker criteria (1981), the discriminant value is assessed through the heterotrait - monotrait ratio, referred to as the index HTMT The HTMT coefficient is the ratio of the correlation coefficient between features (between-trait correlations) to the correlation coefficient within the features (within traits)

31

The scale achieves discriminant value when the HTMT index is less than | and preferably less than 0.9 (Henseler et al., 2015)

3.4 Assessing Structural model

Structural model assessment in PLS-SEM focuses on evaluating the significance and relevance of path coefficients, followed by the model’s explanatory and predictive power Structural model assessment - A systematic process is also followed to evaluate a structural model in PLS-SEM with the following steps: (1) evaluate multicollinearity between the independent variable constructs of the structural model; (2) examine the size and statistical significance of the path coefficients; assess in-sample prediction of the dependent constructs based on (3) the R’ of the endogenous variable(s), (4) the effect size (f?), and (5) the predictive relevance (Q’); and (6) evaluate the out-of-sample predictive validity using PLS predict (Shmueli et al., 2019) If the recommended rules of thumb are met for measurement and structural models, then the researcher is ready to report the findings For a more detailed description of the structural model assessment steps, see Hair Jr et al (2020)

3.4,1 Assessing Multicollinearity

Multicollinearity occurs when two or more independent variables have a high correlation with one another in a regression model, which makes it difficult to determine the individual effect of each independent variable on the dependent variable

Multicollinearity can occur due to poorly designed experiments, highly observational data, creating new variables that are dependent on other variables, including identical variables in the dataset, inaccurate use of dummy variables, or insufficient data

One method to detect multicollinearity is to calculate the variance inflation factor (VIF) for each independent variable, and a VIF value greater than 1.5 indicates multicollinearity This is captured by the VIF, which is denoted below:

1 1—R?2

component analysis to reduce the number of variables while retaining most of the information

3.4,2, Relationship in structural model

Due to the fact that PLS-SEM does not require that the data be regularly dispersed The parameterized tests used in the regression analysis to determine whether or not the outer weight, outer loading, and path coefficients are statistically significant cannot be performed because there is no normal distribution In order to verify the significance level, PLS-SEM employs a statistically significant coefficient based on its standard error, which is acquired by bootstrapping

According to Hair et al (2018), a return enlarged sample of about 5,000 samples was suggested We can get the experimental t-value and p-value for each of the path systems in the structural model using the bootstrap standard error The test is statistically significant at the 5% level with a t-value > 1.96

3.4,3 Assessing Coefficient of determination (R*)

The fundamental factor in this evaluation is the coefficient of determination (R’) of the endogenous latent variables R? is computed as the square of the correlation between the predicted and observed values of the particular dependent research variable R? varies from 0

to 1, with higher R-squared values indicating a better fit In PLS path models, R? values of 0.67, 0.33, and 0.19 are interpreted as substantial, moderate, and weak, respectively (Henseler

et al., 2009)

3.4.4, Assessing Effect Size (f}

To assess the magnitude of each effect in the path model, one can utilize Cohen’s (1988) f? This measure determines the effect size by comparing the increase in explained variance (R?) with the proportion of variance of the endogenous latent variable that remains unexplained Values of f? such as 0.02, 0.15, and 0.35 denote small, medium, and large effects, respectively (Cohen, 1988)

3.4.5 Assessing the relevance of (@)

The primary method for assessing predictive relevance is Stone-Geisser's Q?, introduced

by Stone (1974) and Geisser (1975), typically evaluated through blindfolding procedures as outlined by Tenenhaus et al (2005) According to the Stone—Geisser criterion, the model should be capable of predicting the indicators of the endogenous latent variable This technique combines elements of function fitting and cross-validation Chin (1998) emphasizes

33

that predicting observable or potentially observable variables holds more significance than estimating parameters of artificial constructs

3.5 Measurement scale

The research papers were evaluated using 5-point scales based on prior literature and translated into Vietnamese, the official language of the current research context On a 5-point scale, we scored each item from | (strongly disagree) to 5 (strongly agree)

Table 3.1 Measurement scales

Construct Source Items Questions Adjusted Questions

EC1 | I’m very concerned I’m very concerned about current about current environmental environmental pollution in China and | pollution in Vietnam its impact on health and its impact on

health

EC2 | Automobile exhaust Automobile exhaust emission is one of the | emission is one of the Environmenta primary sources of air | primary sources of air

1 Concern Transportation pollution pollution

(Wu, J et al., Research Part EC3 | I have the Ihave the

2018) F responsibility to adopt | responsibility to adopt

a low-carbon travel a low-carbon travel mode mode

HC1 | I reflect about my I reflect about my health a lot health a lot

Health International HC2 | I’m very self- I’m very self- Consciousness | journal of conscious about my conscious about my (Shimoda, A et | environmental health health

al., 2020) health HC3 | I’m generally attentive | I’m generally attentive

research to my inner feelings to my inner feelings

about my health about my health PBC1 | I believe I have the I believe I have the ability to purchase a ability to adopt hybrid car electric vehicles

34