BÁO CÁO TÔNG KẾTĐÈ TÀI NGHIÊN cứu KHOA HỌC THAM GIA XÉT GIÃI THƯỞNG ‘’NHÀ NGHIÊN CƯU TRẺ UEH” NĂM 2024 EXPLORING FACTORS AFFECTING THE INTENTION OF USING ELECTRIC MOTORBIKE TAXI SERVICE
INTRODUCTION
Research Background
Pollution and climate change are primary drivers of natural disasters and epidemics, highlighting the urgent need for environmental protection as a key component of sustainable development (The United Nations, 1992) As nations strive for sustainable progress, countries like Vietnam are increasingly prioritizing environmental conservation efforts.
Transportation is a significant contributor to global warming, responsible for over 16% of greenhouse gas emissions (Ritchie, H., 2020) To mitigate climate change, a swift reduction in vehicle emissions is essential Electric vehicles (EVs) offer a promising solution by eliminating exhaust emissions, thus helping to combat air pollution and global warming However, challenges such as long charging times and insufficient charging infrastructure hinder their market penetration The disparity between the number of public EV charging stations and gasoline stations is notable, with EV stations being far fewer, and charging times considerably longer than refueling gasoline vehicles (Jung, J., & Chow, J Y., 2019) Without an expansion of public charging stations, long-distance travel for EV users will be limited, reducing their appeal for households lacking private garages (Hall, D., & Lutsey, N., 2017) Despite these hurdles, research indicates potential solutions; Gupta et al (2021) advocate for a centralized approach to charging station allocation, enhancing efficiency and accessibility, while Mortimer et al (2022) propose a data-driven method to place charging stations based on real-world usage and popular locations The EV industry is also innovating with hybrid, hydrogen-powered vehicles, and solar-powered charging stations, alongside advancements in battery technology, making EVs a viable and eco-friendly alternative to traditional fossil-fuel cars (Nguyen, M H., 2023).
Electric vehicles (EVs) present a sustainable alternative to fossil fuel-powered vehicles, significantly reducing air pollution Unlike traditional vehicles that depend on non-renewable fossil fuels, EVs utilize electricity generated from renewable sources such as solar and wind energy With the transportation sector consuming nearly 49% of global oil demand, the depletion of oil and gas resources is a growing concern, projected to occur by 2038 due to current consumption trends This looming crisis threatens the global economy, likely resulting in soaring fuel prices and creating challenges for both consumers and energy providers worldwide.
Governments worldwide, including Vietnam, are implementing measures to reduce the use of internal combustion engine vehicles and promote electric vehicles The European Commission announced a ban on all internal combustion engine vehicles by 2035 as part of its "Fit for 55" legislation, aiming for a carbon-neutral continent Brussels is leading this initiative with plans to prohibit diesel and gasoline vehicles by 2030 and 2035, respectively, while introducing regulations to encourage electric vehicle adoption, such as increasing gasoline car prices and penalizing carbon-emitting manufacturers Similarly, Vietnam's Decision No 876/QD-TTg aims for a green transportation system with zero net greenhouse gas emissions by 2050, focusing on enhancing electric charging infrastructure and providing green energy nationwide to meet consumer needs The primary motivation for this shift to electric vehicles is the urgent need to mitigate environmental impacts.
Electric motorcycles have been present in Vietnam since the 2010s, while electric cars have only recently entered the domestic market VinFast currently dominates this sector, having established an extensive network of nearly 10,000 charging stations across the country, significantly enhancing the accessibility of electric vehicles for consumers.
The Vietnamese market is witnessing a surge in electric vehicles, with brands like Hyundai, Mercedes, and various Chinese manufacturers introducing their models This growth in the electric vehicle sector is paralleled by the rapid development of motorbike taxi services, notably marked by the emergence of VinFast's Green SM service As Vietnam's first dedicated electric vehicle booking service, Green SM operates exclusively with VinFast's odorless and noiseless electric cars and motorcycles, promoting user health and environmental sustainability.
SM has officially launched in major cities in Vietnam, with Ho Chi Minh City being one of the earliest locations to introduce the Green SM electric vehicle service nationwide.
Ho Chi Minh City, a bustling megacity in Southeast Asia, boasts a high population density with millions migrating for work and life, fostering a vibrant cultural and economic landscape This dynamic environment promotes investment opportunities across various sectors, making the city's economy one of the most developed in Vietnam The rise of motorbike taxi services, exemplified by successful companies like Grab, Gojek, and Be, reflects the increasing demand for fast, safe, and affordable transportation among the growing student and migrant worker population However, despite this economic progress, the city grapples with significant air pollution challenges, particularly due to high levels of fine dust.
The alarming levels of PM2.5 pollution, which exceed the WHO's annual air quality guideline by 4.4 times, raise significant health concerns As awareness of these potential health risks grows, individuals are increasingly seeking eco-friendly transportation options to reduce air pollution This trend presents a unique opportunity for electric motorbike taxi services, which are steadily gaining market share Consequently, we have selected Ho Chi Minh City as the focal point for our study on the factors influencing the intention to use electric motorbike taxi services.
VinFast, recognized as the first company to deliver electric vehicles (EVs) to local customers, demonstrates significant financial potential (Doll, 2021) The Green - Smart - Mobility (GSM) Joint Stock Company, established by VinFast, is Vietnam's pioneering taxi service provider utilizing electric cars (VNA, 2023) As stated on the GSM website, their mission is to promote the use of electric vehicles among the public, thereby raising awareness about the sustainability of green transportation This electric taxi service is celebrated for providing smooth, noise-free, and emission-free rides, earning high praise from users since its founding on March 6th.
In 2023, GSM has rapidly gained traction in the market, amassing 6 million customers in just five months and covering half of Vietnam's provinces (Capital Security newspaper, 2023) The company has also extended its reach to Laos, aligning with its "Go Green Global" strategy to position itself as a leading international electric mobility service provider and to foster the adoption of green mobility across borders (KPL, 2023).
Numerous studies investigate various factors influencing the adoption of electric vehicles (EVs), including the distance an EV can travel on a single charge and the distribution of charging infrastructure (Wahab & Jiang, 2019; Merkisz-Guranowska & Maciejewski, 2015) Additionally, research highlights the cost implications associated with EV usage (Slowik & Lutsey, 2018; Cano et al., 2018; Jonatan et al., 2018; Bockarjova & Steg, 2014) and examines the environmental impacts of electric vehicle adoption (Trin Thananusak et al., 2017; Wujin Chu et al., 2019) Furthermore, studies also delve into public attitudes towards electric vehicles (Oliveira et al., 2022; Wang et al., 2016; Gunawan et al., 2022).
Research on the intention to use electric taxi services in Europe and America is limited, primarily focusing on the acceptance and trade of electric vehicles In contrast, Asian markets predominantly study motorbike taxi services, such as Grab and Gojek, which have established loyal customer bases This study aims to explore the factors influencing young people's intentions to use electric taxi services in Ho Chi Minh City, particularly through the case study of Green SM By analyzing customer behaviors and their intentions regarding electric motorbike taxi services, our findings contribute valuable insights to the ongoing discussions about sustainable transportation and environmental conservation.
The Previous Study
To reduce fuel consumption in transportation, replacing internal combustion engines with alternatives like electric or hydrogen vehicles is essential (Knittel, C R., 2012) However, electric vehicles encounter market penetration challenges, primarily due to concerns about their range and the insufficient infrastructure for recharging Research indicates that the availability of electric charging stations and the driving range of electric vehicles significantly impact consumer adoption According to Houde (2012), the most beneficial refueling stations for potential buyers are those located along common driving routes and at destinations rather than solely at home The study highlights the existing electric charging network's role in shaping Swiss citizens' perceptions of electric vehicles, emphasizing the importance of strategically positioned charging stations along routes and their connectivity to starting and ending journey points (Li, J., 2019).
Li (2019) emphasizes the importance of fuel infrastructure, particularly electric charging stations, in enhancing mobility for electric vehicle drivers The travel distance of an electric vehicle on a fully charged battery generally improves with a larger battery size However, to achieve a greater range, the larger and heavier battery consumes more energy, ultimately increasing the vehicle's cost.
A study by Wahab et al (2019) employs Logit and Probit models to identify factors affecting electric motorcycle adoption in Northern Ghana, based on interviews with 627 motorcycle owners over two months The research examines motorcyclists' perceptions of technical specifications, including charging times, battery lifespan, performance, pricing, driving range, age, and monthly income Findings indicate that key factors such as "Maximum distance," "Performance of the electric motorcycle," and "High usage" significantly influence motorcyclists' willingness to adopt electric motorcycles.
Merkisz-Guranowska et al (2015) highlight the advancements in electric taxi fleets across various cities, focusing on the potential implementation of electric taxicabs in Poznan, Poland, a city with a population of 600,000 The study details the selection of vehicles, the calculation of operating costs for taxicabs, and the assessment of expenses related to the necessary charging infrastructure.
Recent advancements in battery technology, such as IBM's Li-Air battery, can provide an impressive range of up to 800 km, matching the capabilities of the Mercedes EQS electric vehicle available in Vietnam Additionally, research has focused on enhancing electric vehicle charging infrastructure, with studies proposing optimal allocation policies for public charging stations (Gupta et al., 2021) and suggesting deployment processes based on actual usage data (Mortimer et al., 2022) Beyond these technological improvements, the price of electric vehicles remains a crucial factor affecting consumer adoption Given that electric vehicles generally have higher upfront costs compared to traditional gasoline cars, potential buyers need to recognize economic advantages such as battery efficiency, operating range, purchase price, charging time, and overall usage costs to justify their investment (Axsen & Kurani, 2013).
A study by Slowik et al (2018) examined various factors that influence consumers' intentions to adopt electric vehicles in the United States, highlighting the significance of awareness, government commitment, sales, customer service, research and development, financing, battery technology, development technology, and user training The findings indicate that competitive financial costs of electric vehicles in comparison to gasoline vehicles play a crucial role in advancing the electric vehicle industry.
Cano et al (2018) conducted a comparative study on electric vehicle consumption habits among Americans and citizens of emerging nations, including China, India, Brazil, and Indonesia The findings indicate that the United States has the highest willingness to pay for electric vehicles with extended driving ranges, reflecting a strong preference for higher battery capacities Furthermore, the study highlights that both the U.S and China exhibit a low negative index when it comes to price increases, suggesting that citizens in these countries have higher incomes and greater financial capability to afford electric vehicles.
A survey conducted by Jonatan et al (2018) involving 1,248 car owners across France, Germany, Italy, Poland, Spain, and the United Kingdom reveals that price is the most significant factor influencing the intention to adopt electric vehicles in Europe Furthermore, the majority of respondents view government incentives as a crucial element in their decision-making process for purchasing electric cars However, the effectiveness of these incentives can differ across countries due to varying socioeconomic characteristics among consumers.
Bockarjova and Steg (2014) highlight that the Protection Motivation Theory effectively models factors influencing electric vehicle adoption among Dutch consumers Their research indicates that individuals are more inclined to consider electric vehicles when they recognize the serious negative impacts of conventional vehicles and believe that electric vehicles can alleviate these issues However, the primary obstacles to adopting electric vehicles include the high financial and non-financial costs associated with them, alongside the perceived advantages of conventional vehicles.
However, some other studies also indicate that consumers are willing to accept higher costs if the electric vehicle is environmentally friendly and meets their criteria
A study conducted by Trin Thananusak et al (2017) on the factors influencing the intention to purchase electric vehicles in Thailand surveyed 149 experienced adult consumers, resulting in 102 valid responses The analysis revealed three key variables positively impacting Thai citizens' intentions to buy electric vehicles: performance factors, environmental concerns, and price-premium The findings indicate that Thai consumers are willing to pay a higher price for electric vehicles that are environmentally friendly and meet their established criteria.
Figure 1: Conceptual framwork (Trin Thananusak et al., 2017)
A study by Chu et al (2019) involving 177 electric vehicle users in South Korea and 204 in Beijing, China, revealed that Chinese respondents exhibited a higher level of environmental concern than their South Korean counterparts This heightened awareness of environmental issues is identified as the primary motivator for Chinese consumers purchasing electric vehicles Despite both countries grappling with air quality challenges, the significantly greater emphasis on environmental concerns among Chinese citizens is reflected in the higher standardized beta coefficient for this factor in surveys conducted in China.
Environmental concerns have minimal influence on South Koreans' acceptance of electric vehicles, with lower fuel costs and government subsidies being the primary motivators This study offers valuable insights for businesses targeting electric vehicle consumers in China and South Korea Companies should focus on promoting the technological advancements of electric vehicles, as South Koreans prioritize innovation in their satisfaction In contrast, manufacturers in China should emphasize the substantial environmental benefits associated with electric vehicles.
A segment of the population exhibits a strong emotional connection and positive attitudes towards electric cars, valuing their innovative features and modern technology These individuals are open to the concept of "learning by driving" and accept the imperfections associated with electric vehicles (Abenoza et al., 2017) A study by Oliveira et al (2022) explored various dimensions such as attitude, emotions, subjective norms, and perceived behavioral control among Brazilian drivers Conducted via Google Forms from April to May 2019, the survey included 488 participants The findings revealed that emotions significantly influence attitudes towards electric cars, with emotional factors being the most impactful, while perceived behavioral control showed no significant effect, and subjective norms did not meet the criteria for validation.
A survey conducted by Wang et al (2016) explored the attitudes of Chinese consumers towards electric vehicles, focusing on key factors from the Theory of Planned Behavior: subjective norms, perceived behavioral control, and personal moral norms The findings revealed that consumer environmental concern indirectly influenced the intention to adopt electric vehicles, showing a significant positive relationship with attitudes toward these vehicles, perceived behavioral control, subjective norms, and personal moral norms, ultimately enhancing the intention to purchase electric vehicles.
The study of Gunawan at al (2022) examine the factors influencing the intention to use electric vehicles in fifteen major cities in Indonesia The survey was conducted on
526 individuals from July to December 2021 The survey participants were Indonesian residents aged 17 and above with a driver's license The main dimensions included attitude, perceived behavior control, and Subjective norms.
Figure 2: Study Conceptual Model (Indra Gunawan et al., 2022)
Intention to Use Electric Vehicles
The study found that all dimensions had an impact, with the attitude dimension having the strongest impact This demonstrates people's belief in the influence of electric vehicles on Indonesia.
In Vietnam, various studies have utilized the Theory of Planned Behavior (TPB) to investigate consumers' intentions to adopt electric vehicles, confirming its relevance and effectiveness For instance, Phuong (2023) conducted an online survey with 291 respondents, identifying five key factors influencing electric vehicle purchase intentions: Attitude, Perceived Behavioral Control, Electric Vehicle Awareness, Subjective Norms, and Environmental Concern The findings indicated that Attitude significantly impacts the intention to purchase electric vehicles in Ho Chi Minh City, followed by Perceived Behavioral Control and Subjective Norms, while Environmental Concern had the least influence Additionally, other studies have expanded on TPB by exploring additional factors affecting electric vehicle adoption Pham et al surveyed over 300 high school students in Hanoi, identifying seven determinants for using electric bicycles, including Convenience of Use and Environmental Friendliness Similarly, Tran et al (2021) examined the intention to use VinFast electric motorcycles among 153 residents of Ho Chi Minh City, highlighting Appeal of Alternative Transportation, Attitude, and Environmental Awareness as the primary influencing factors.
Research Objectives
This research aims to explore the factors influencing young people's intention to use electric motorbike taxi services in Ho Chi Minh City Specifically, it seeks to identify and assess the strength of these factors affecting usage intention Additionally, the study proposes management strategies to enhance the adoption of electric vehicle services among students in the city.
Research Questions
This study aims to explore the factors influencing young people's intention to use electric taxi services in Ho Chi Minh City To achieve this, three specific research questions have been formulated based on the research background and objectives.
RQ1: What factors impact on the Usage intention of electric taxi services of young people who are living in Ho Chi Minh city?
RQ2: Whether each factor impacts the Usage intention of electric taxi services and the strength of its influence?
RQ3: What should businesses do to increase the intention to use electric vehicle services among students in Ho Chi Minh City?
Research Scope
(1) Research subjects: Students in Ho Chi Minh City.
(2) Survey time frame: From January 17th, 2024 to February 8th , 2024.
Research Methodology
Data was gathered via online surveys targeting university students in Ho Chi Minh City, utilizing both qualitative and quantitative research methods The research team employed SPSS 27.0 software to analyze the data through four key steps: descriptive statistics, reliability testing with Cronbach's Alpha coefficient, exploratory factor analysis (EFA), and linear regression analysis.
FUNDAMENTAL THEORIES AND RESEARCH MODEL
Electric taxi service
Electric Vehicles (EVs) are defined as vehicles equipped with at least one electric motor for propulsion Battery Electric Vehicles (BEVs) utilize only batteries and an electric motor, while Fuel Cell Electric Vehicles (FCEVs) combine fuel cells and batteries with an electric motor Additionally, Hybrid Electric Vehicles (HEVs) employ both batteries and liquid fuel, using both an engine and an electric motor for propulsion.
Zhou et al (2015) gives the definition for EVs and other types of electric vehicles:
- EVs: any vehicle that uses energy drawn from the electric grid and stored onboard the vehicle for some or all of its propulsion (i.e., PHEVs and BEVs).
- PHEVs: vehicles that use energy stored from the grid but also have an internal combustion engine to provide driving range and vehicle power.
- BEVs: vehicles that use energy stored from the grid exclusively.
- HEVs: vehicles that generate all of their power onboard the vehicle
This study defines "electric vehicle" specifically as pure electric vehicles, including Plug-in Electric Vehicles (PEVs) and Battery Electric Vehicles (BEVs), which operate solely on electricity without any alternative energy sources.
The evolution of electric vehicle technology has entered its third century and is poised for swift advancements in the foreseeable future (Larminie, J., & Lowry, J.,
The electric car market presents unique challenges and opportunities compared to conventional automobiles, particularly in terms of service aspects that influence consumer purchasing decisions Effective management of the "Product Service System" (PSS) is crucial, especially in emerging markets like electric vehicles, where companies must innovate their product-service offerings Despite the growing importance of PSS in the electric car sector, there remains a scarcity of scholarly literature addressing this intersection Cherubini et al (2015) highlight the need for research in this area, focusing on the integration of product-service systems with the electric vehicle market to identify Critical Success Factors (CSFs) that can enhance the adoption of electric cars.
In Viet Nam, Green SM is the first pure electric motorbike taxi service brand, offering unique transportation experiences: Green SM Taxi standard taxi service, Green
SM Luxury premium taxi service, and flexible electric scooter transportation service Green SM Bike.
Green SM
According to the GSM website, Green SM is the brand name of the first electric taxi service from the Republic of Vietnam Developed by the Vingroup conglomerate, Green
Green SM is a pioneering electric taxi and car rental service in Vietnam, operating a fleet of 10,000 cars and 100,000 electric motorcycles As the first all-electric taxi service in the country, it distinguishes itself from traditional gasoline-powered taxis by prioritizing environmental sustainability The company's name reflects its dedication to clean energy and modern technology, with "Green" symbolizing the environment and "Blue" representing intelligence Overall, Green SM's commitment to electric vehicles marks a significant advancement in the transportation industry, serving as a model for future innovations in sustainable mobility.
Fundamental Theories
The Theory of Planned Behavior (TPB) is a significant framework in understanding customer purchasing behavior, building upon the earlier Theory of Reasoned Action Developed by Ajzen, TPB introduces "Perceived Behavioral Control" as a crucial third factor, alongside "Attitude" and "Subjective Norm," that influences consumer decision-making.
Figure 3: Theory of Planned Behavior (Ajzen, 1991)
The Theory of Planned Behavior, proposed by Ajzen in 1991, emphasizes that purchasing intentions significantly influence consumer behavior It identifies three key factors: social influence, which pertains to societal perceptions of purchasing behavior; individual factors, reflecting a person's positive or negative attitudes toward consumption; and perceived behavioral control, which relates to an individual's self-awareness and ability to execute the purchasing action.
Grounded in the Theory of Planned Behavior, our study integrates three key factors—Attitude, Subjective Norm, and Perceived Behavioral Control—to examine their influence on the intention to adopt electric vehicle services.
2.3.2 Theory of Tanwir and Hamzah
The study of Tanwir and Hamzah (2020) presents four hypotheses regarding the electric vehicle purchasing behavior of Malaysian citizens.
Figure 4: Theory of Tanwir and Hamzah (2020)
This study explores the key factors that impact consumers' decisions to purchase electric vehicles, highlighting the significance of "Green purchase awareness." This awareness encompasses consumers' knowledge and understanding of electric vehicles and the brands associated with them.
User preferences play a vital role in the selection of electric vehicles, significantly shaping attitudes toward their adoption Acceptance of electric vehicles often hinges on various evaluations and opinions, which can be both positive and negative, concerning environmentally friendly practices Furthermore, the distinct effects of perception and attitude are critical in influencing decision-making and purchasing behavior related to electric vehicles.
An additional variable introduced in this study is "Environmental concern"
Environmental concern encompasses an individual's awareness and worry about environmental issues, particularly the adverse effects of human actions on nature Research indicates that individuals with heightened environmental concern are more likely to consider purchasing electric cars, as their awareness of environmental impacts influences their buying decisions.
Therefore, the authors' study will utilize the theory proposed by Tanwir and Hamzah
(2020) as a foundation, incorporating the factor of "Environmental concern" into the research model to further analyze its impact on the intention to use electric vehicle services.
Research Hypotheses
Attitude, as defined by Fishbein and Ajzen (1975), encompasses an individual's positive or negative evaluations and emotions regarding specific behaviors Research has consistently shown that a favorable attitude towards an action significantly enhances the intention to engage in that behavior (Chan R Y K., 2001; Vermeir, I., & Verbeke, W., 2004) Numerous studies highlight the strong correlation between consumers' attitudes and their behavioral intentions towards green consumption across various cultural contexts, including Asia, the Americas, and Europe, as well as across different product categories, such as green products (Yadav, R., & Pathak, G S., 2016) and electric vehicles (Huang, X., & others).
Ge, J., 2019) Therefore, the first hypothesis is proposed as follows:
Hypothesis Hl: "Attitude has a positive effect (+) on the intention to use Green
Subjective norm refers to an individual's perception of societal attitudes toward a specific action, such as using electric vehicle services (Ajzen, 1991) When individuals believe that others view the use of these services positively, they are more likely to intend to use them Research by Bamberg (2003) has shown a strong correlation between subjective norms and the intention to consume green products Thus, we propose the following hypothesis:
Hypothesis H2: "Subjective norm has a positive effect (+) on the intention to use
Perceived behavioral control, introduced by Ajzen in 1991 to enhance the Theory of Reasoned Action, refers to an individual's perception of the ease or difficulty in performing a specific behavior, influenced by available resources and opportunities Ajzen asserts that this perception directly affects the likelihood of engaging in the behavior, and accurate self-assessment of control can predict actual behavior Previous research has demonstrated that perceived behavioral control positively impacts behavioral intentions, leading to the formulation of the third hypothesis.
Hypothesis H3: "Perceived behavioral control has a positive effect (+) on the intention to use Green SM."
Environmental concern significantly influences consumer behavior, particularly regarding green consumption intentions Research indicates that individuals who prioritize environmental quality are more inclined to invest in eco-friendly products, such as electric vehicles, as they perceive the long-term benefits to outweigh the initial costs This leads to the hypothesis that heightened environmental awareness correlates with a greater willingness to pay for sustainable options.
Hypothesis H4: "Environmental concern has a positive effect (+) on the intention to use Green SM."
Price value is the monetary expression of the value of a commodity, meaning the amount of money to be paid for a commodity, service, or asset In the study of Thanh
In 2020, regression analysis indicated that the "Price value" factor significantly influences customers' intention to adopt technology-based car services The findings recommend enhancing promotional strategies, such as discounts and loyalty programs, to improve price competitiveness For students in Ho Chi Minh City, the cost of Green SM electric vehicle services plays a crucial role in their decision-making process, especially given the availability of other more affordable technology-based car services Consequently, the relevance of "Price value" is particularly pronounced for students considering the use of Green SM electric vehicle services in Ho Chi Minh City.
Hypothesis H5: "Price value has a positive effect (+) on the intention to use Green
Research Model
RESEARCH METHODOLOGY
Research Process
This research employed a quantitative approach to assess the reliability of each variable and validate the independent and dependent factors By utilizing this method, the study effectively tested hypotheses and models while measuring the strength of each independent factor, including the "Subjective norm."
The study investigates how "Attitude," "Perceived Behavioral Control," "Environmental Concern," and "Price Value" influence the intention to use Green Social Media (SM) Qualitative data was collected using non-probability sampling, and the analysis was conducted with SPSS 27.0 software A research process was designed to identify the factors affecting the usage intention of Green SM and to assess the strength of each factor's impact, as illustrated in Figure 6.
Research Method
First, the authors conducted qualitative research methods as follows:
(1) Review of theoretical foundations and related research articles, thereby proposing a research model and preliminary scale.
(2) Discussion to consider and adjust the use of language, content for appropriateness to ensure that the survey respondents understand correctly about the questionnaire.
(3) Develop an official scale to proceed with quantitative research.
Following qualitative research, quantitative research will utilize consumer survey data The survey responses will be encoded and analyzed with SPSS software, involving steps like sample characteristic description, Cronbach’s Alpha testing, difference testing, exploratory factor analysis (EFA), Pearson correlation analysis, and linear regression analysis.
Measurement Scales
3.3.1 Measurement scales for independent factors
This research identifies key independent factors influencing the usage intention of Green Social Media (SM) among young people in Ho Chi Minh City, including Attitude, Subjective Norm, Perceived Behavioral Control, Environmental Concern, and Price Value To assess the impact of each factor, measurement variables were established based on prior studies, as detailed in Table 1.
Table 1: Measurement variables for independent for independent factors
1 If people around me use Green SM electric motorbike taxi service, this will prompt me to use.
2 Most people who are important to me think I should use Green SM electric
Safian et al., 2023 motorbike taxi service in the near future.
3 News media propaganda will prompt me to use Green
SM electric motorbike taxi service.
4 If I use Green SM electric motorbike taxi service, it would be consistent with the trend of social development.
1.1 think it is very necessary to use Green SM electric motorbike taxi service.
2 I think using Green SM electric motorbike taxi service is a good choice.
3 I support the country in introducing more policies to encourage individuals to use Green SM electric motorbike taxi service.
4.1 think Green SM electric motorbike taxi service satisfies my demand for using a ride-hailing service.
1.1 think Green SM electric motorbike taxi service are user-friendly.
2 I have resources to use Green SM electric motorbike taxi service.
3.1 have time to use Green
SM electric motorbike taxi service.
4 Whether or not I use Green
SM electric motorbike taxi service is completely up to me.
5.1 have sufficient knowledge to use Green SM electric motorbike taxi service.
1.1 think environmental issues are becoming increasingly serious in recent years.
Wang, s., Fan, J., Zhao, D., Yang, s., & Fu, Y., 2016
2 I think we should have the responsibility to protect the environment.
3 I think the condition of the environment affects my health quality Yegin, T., &
EC4 4.1 am concerned about environmental issues.
5 I think humans are seriously abusing the environment
1 Using Green SM electric motorbike taxi service is reasonably priced.
2 Green SM electric motorbike taxi service offers prices that are suitable for the quality of service.
3.3.2 Measurement scales for dependent factors
3 Green SM electric motorbike taxi service generates more value than the cost incurred.
4 Green SM electric motorbike taxi service has the suitable price for my needs.
Based on previous studies, the four variables for dependent factor - The Usage intention of Green SM are showed in Table 2.
Table 2: Measurement variables for independent for dependent factors
1 I will use Green SM electric motorbike taxi service in the future.
UI2 2 If the Green SM electric 2019 motorbike taxi service is good, I would recommend friends around me to use.
UI3 3 If I use a ride-hailing service in the future, I will think of choosing Green SM electric motorbike taxi service.
UI4 4.1 hope Green SM electric motorbike taxi service will be more popular in the future.
Questionnaire Design
The content of the survey questionnaire consists of three main sections as follows:
(1) General Information: Questions used for sampling purposes, aiding in eliminating observations that do not meet the study's scope requirements from the data to be processed.
(2) Personal Information: Questions regarding demographic information (gender, courses, frequency of using motorbike taxi services), serving as the database for testing differences in intention among different target groups.
Gender: 03 values (Male, female, other)
Course: 05 values (First year, second year, third year, fourth year, others)
Frequency of using motorbike taxi services: 05 values (Never, rarely, occasionally, frequently, always)
The survey employs a 6-level Likert scale ranging from "Strongly Disagree" to "Strongly Agree" to collect respondents' opinions and perspectives, forming a foundational database for quantitative research The levels are systematically arranged from 1 to 6 to ensure clarity and consistency in responses.
Data Collection Method
The study focuses on students residing in Ho Chi Minh City, as they are well-versed in new technologies and proficient in using digital devices and social media This demographic is particularly attentive to environmental concerns and the demand for eco-friendly products Given that Ho Chi Minh City serves as Vietnam's economic hub, there is a significant need for electric taxi services among this group.
The research employed a non-probability sampling method, conducting an online survey in Ho Chi Minh City This approach allowed for easier access to the target demographic, with survey forms distributed and collected through Google Forms.
The reliability and validity of variables were assessed using Cronbach's Alpha and Exploratory Factor Analysis (EFA) Following this, multiple regression analysis was performed to evaluate the hypotheses It is essential to have a sufficiently large sample size for statistical analysis; according to Hair et al (2010), the minimum sample size for EFA is 50, although a sample size of 100 or more is preferable, with at least 5 observations per measured variable.
In the study involving 25 observed variables, a minimum sample size of 125 observations is necessary to fulfill quantitative analysis requirements According to Tabachnick and Fidell (2001), for multiple regression tests, the minimum sample size should be calculated as n > 50 + 8k, where k represents the number of independent factors With five independent factors in this study, the minimum sample size would be 90 However, to account for potential exclusions of inappropriate responses and to enhance the reliability and representativeness of the survey, the anticipated initial sample size is set at 200.
Data Analysis Method
Data collected from the sample were analyzed to examine the relationships between independent and dependent variables The process began with coding each variable, followed by assessing reliability using Cronbach’s Alpha, validating the data through Exploratory Factor Analysis (EFA), and testing hypotheses with multiple regression analysis.
To test the suitability of path model, the generalized squared multiple correlations was used (Pedhazur, E J., & Kellinger, F N., 1982).
To investigate the connection between demographic variables and the intention to use Green Social Media (SM), a One-way ANOVA was conducted The analysis was performed using SPSS 27.0 software.
To assess the reliability of variables, Cronbach’s Alpha was employed, with an ideal range of 0.75 to 0.95 for optimal results According to Nunnally and Bernstein (1994), a Cronbach’s Alpha of 0.6 or higher is deemed acceptable Additionally, a corrected item-total correlation value exceeding 0.3 indicates that the variables are suitable for reliability testing.
To assess the validity of independent and dependent factors, Exploratory Factor Analysis (EFA) was utilized, employing the varimax rotation method The EFA was conducted twice: once for the independent factors and once for the dependent factor This analytical technique is designed to condense numerous observed variables into more significant underlying factors, adhering to specific criteria.
- Kaiser-Meyer-Olkin (KMO) Measure: Assessing the adequacy of the factors included; if 0.5 < KMO < 1, the factor analysis is appropriate.
Bartlett's Test of Sphericity assesses the correlation among observed variables within a factor; a significance level (Sig.) of less than 0.05 indicates significant correlations among these variables Additionally, factors are retained in the model if their Eigenvalue is greater than 1, while those with an Eigenvalue less than 1 are excluded.
- Total Variance Explained > 50% indicates the suitability of the EFA.
- The study employs the Vari max orthogonal rotation method to eliminate variables with factor loading < 0.5.
Factor loading reflects the correlation interaction between the factor and the observed variables:
- Factor loading > 0.3: Minimum condition for retaining observed variables.
- Factor loading > 0.5: Observed variables have good statistical significance.
- Factor loading > 0.7: Observed variables have very good statistical significance.
To test the hypotheses in research model, the multiple regressions analysis has been applied to assess the effects of the independent factors on the dependent factor.
- Perform Pearson correlation analysis to examine the linear correlation between dependent and independent variables and to guard against multicollinearity.
+ Pearson correlation ranges from -1 to 1.
+ If Sig < 0.05, there is a correlation between the independent and dependent variables.
- Construct a linear regression model using the ordinary least squares method The Enter variable selection method is employed Adjusted R-squared is used to determine the model's fit.
- ANOVA test is used to assess the model's fit If the F-statistic has Sig < 0.05, it can be determined which independent variables explain the variation in the dependent variable.
The coefficient B indicates how much the dependent variable changes with a one-unit increase in the independent variable, assuming other variables remain constant; thus, a higher adjusted B coefficient signifies a stronger impact on the dependent variable.
- Conduct regression assumption violation checks:
4- Use scatterplots to check the linear relationship assumption between independent and dependent variables and constant residual variance.
+ Use P-P plots or histograms to check residual assumption.
+ Use scatterplots of residuals and predicted values to check the assumption of constant variance of the dependent variable.
+ Use the Durbin-Watson statistic to check the assumption of no autocorrelation between residuals Use the Variance Inflation Factor (VIF) to check for multicollinearity; if VIF > 2, multicollinearity may occur.
One-way ANOVA test will be applied to examine the mean differences among demographic variables with three or more attributes.
- If Sig Levene statistic > 0.05, the variances among groups are not different, then we proceed to examine the Sig in the ANOVA test table.
- If Sig < 0.05, there is a statistically significant difference in intention among different groups.
DATA ANALYSIS AND RESULTS
Study sample description
Conducting descriptive statistics is essential for analyzing data, as it offers a comprehensive overview of the dataset's key characteristics The research team collected 407 survey responses using Google Forms, but identified 17 invalid samples due to respondents lacking prior experience with ride-hailing services or knowledge about Green SM service After encoding and verifying the data, 390 valid samples were analyzed using SPSS software, with the statistical results presented through charts and detailed information.
Table 3: Genders and Courses Demographics
Table 4: Living Place Distribution of Survey Respondents
Places in Ho Chi Minh City Frequency Percent
Table 5: Frequency of Green SM Service Usage
Green SM Using Frequency Frequency Percent
All the collected data were obtained from students currently residing in Ho Chi Minh City who have used ride-sharing services and have experienced Green SM electric vehicle service.
The gender distribution among participants shows a slight imbalance, with 179 males (45.9%) and 203 females (52.1%), while 8 individuals (2.1%) identified as "others." This distribution is in line with societal norms, indicating an insignificant gender disparity in the study.
The distribution of academic year representation among respondents is fairly balanced, with first-year students making up the largest group at 32.6% (127 students) Second-year students account for 21.8% (85 students), while third-year students represent 18.2% (71 students) and fourth-year students comprise 23.3% (91 students) Additionally, students beyond the fourth year constitute a small percentage at 4.1% (16 students), ensuring no specific group is overly represented.
The survey results indicate extensive coverage across central Ho Chi Minh City, with participants distributed among various districts District 1 and District 10 have the highest representation, comprising 14.6% (57 students) and 12.8% (50 students) of respondents, respectively Additionally, Binh Thanh District contributes a notable 9.5% (37 students), while District 6, District 7, and Binh Tan also have significant participation.
Tan Binh District and Thu Duc City account for an impressive 5% to 6.2% of the student population, while the majority of students live in neighboring districts surrounding central Ho Chi Minh City.
According to the data, a significant portion of students actively utilizes the Green SM service, with 45.4% (177 students) engaging 4 to 5 times per week Additionally, 25.9% (101 students) use the service occasionally, averaging 2 to 3 times per week Notably, 15.4% (60 students) report using the service almost daily, while 13.3% (52 students) have minimal interaction, using it only 1 to 2 times per month.
Data analysis from main questions
Table 6: Descriptive statistical data results
Variables Minimum Maximum Mean Std.
Deviation SN1 If people around me use Green SM electric motorbike taxi service, this will prompt me to buy.
SN2 Most people who are important to me think I should use Green SM electric motorbike taxi service in the near future.
SN3 News media propaganda will prompt me to use Green SM electric motorbike taxi service.
SN4 If I use Green SM electric motorbike taxi service, it would be consistent with the trend of social development.
ATI I think it is very necessary to use
Green SM electric motorbike taxi service.
AT2 I think using Green SM electric motorbike taxi service is a good choice.
AT3 I support the country in introducing more policies to encourage individuals to use Green SM electric motorbike taxi service.
AT4 I think Green SM electric motorbike taxi service satisfies my demand for using a ride-hailing service.
PBC1 I think Green SM electric motorbike taxi service is user-friendly.
PBC2 I have resources to use Green SM electric motorbike taxi service.
PBC3 I have time to use Green SM electric motorbike taxi service.
PBC4 Whether or not I use Green SM electric motorbike taxi service is completely up to me.
PBC5 I have sufficient knowledge to use Green SM electric motorbike taxi service.
ECI I think environmental issues are becoming increasingly serious in recent years.
EC2 I think we should have the responsibility to protect the environment.
EC3 1 think the condition of the environment affects my health quality.
EC4 I am concerned about environmental issues
EC5 I think humans are seriously abusing the environment.
PV1 Green SM electric motorbike taxi service is reasonably priced.
PV2 Green SM electric motorbike taxi service offers prices that are suitable for the quality of service.
PV3 Green SM electric motorbike taxi service generates more value than the cost incurred.
PV4 Green SM electric motorbike taxi service has the suitable price for my needs.
UI 1 I will use Green SM electric motorbike taxi service in the future.
UI2 If the Green SM electric motorbike taxi service is good, I would recommend friends around me to use.
UI3 If I use a ride-hailing service in the future, 1 will think of choosing Green
SM electric motorbike taxi service.
UI4 I hope Green SM electric motorbike taxi service will be more popular in the future.
The survey results indicate that the average values for the majority of observed variables exceed 4.5, suggesting a strong consensus among respondents Additionally, the responses show little variation, reinforcing the overall agreement.
The variable with the lowest average among all variables is “PBC1 - I think Green
The SM electric motorbike taxi service is highly user-friendly, receiving an impressive average rating of 4.27 Notably, two key factors stand out with even higher scores: the recommendation likelihood, with a value of 4.83 for the statement, "If the Green SM electric motorbike taxi service is good, I would recommend it to friends," and the preference for future use, also scoring 4.83 for, "If I use a ride-hailing service in the future, I will consider choosing the Green SM electric motorbike taxi service."
Furthermore, based on the table, the standard deviations range approximately from 1.3 onwards, indicating a considerable fluctuation in survey values This suggests a notable diversity in respondents’ opinions on the survey questions.
The variable "SN1 - If people around me use the Green SM electric motorbike taxi service, this will prompt me to buy" exhibits the smallest standard deviation at 1.310, indicating a high level of consensus among respondents In contrast, two other variables demonstrate the highest standard deviation, reflecting greater variability in opinions.
The Green SM electric motorbike taxi service offers an affordable transportation option, aligning with contemporary social development trends.
4.2.2 Reliability testing (Cronbach’s Alpha value)
Table 7: Cronbach’s Alpha reliability testing first result of “Attitude” scale
Variables Collected Item-Total Correlation Cronbach’s Alpha if Item Deleted
Attitude (AT): Cron bach's Alpha = 0.829
The Cronbach's Alpha value for the "Attitude" scale is 0.829, exceeding the acceptable threshold of 0.6, with corrected item-total correlations for all variables above 0.3 Notably, the Cronbach's Alpha if Item Deleted for the variable "ATI -1 think it is very necessary to use Green SM electric motorbike taxi service" is 0.854, which is higher than the overall scale value To improve the reliability of the scale, the authors have decided to exclude the variable AT 1 and will conduct a second assessment of Cronbach's Alpha.
Table 8: Cronbach’s Alpha reliability testing second result of “Attitude” scale
Variables Corrected Item-Total Correlation Cronbach’s Alpha if Item Deleted Attitude (AT): Cronbach's Alpha = 0.855
The second Cronbach's Alpha analysis for the "Attitude" scale yielded a value of 0.855, indicating strong reliability, as it exceeds the threshold of 0.6 Additionally, the corrected item-total correlations for all variables, AT2, AT3, and AT4, are above 0.3, confirming their acceptability for further factor analysis.
Table 9: Cronbach’s Alpha reliability testing result of “Subjective norm” scale Variables Corrected Item-Total Correlation Cronbach’s Alpha if Item Deleted
Subjective norm (SN): Cwnbach's Alpha = 0.845
The "Subjective Norm" scale exhibits a Cronbach's Alpha value of 0.845, surpassing the acceptable threshold of 0.6, while the corrected item-total correlations for all variables exceed 0.3 Consequently, all variables within this scale are deemed acceptable and will be utilized in the upcoming factor analysis.
Table 10: Cronbach’s Alpha reliability testing first result of “Perceived behavioral control” scale Variables Con*ected Item-Total Correlation Cronbach’s Alpha if Item Deleted
Perceived behavioral control (PBC): Cronbach's Alpha = 0.798
The Cronbach’s Alpha value for the “Perceived Behavioral Control” scale is 0.798, indicating good reliability as it exceeds the threshold of 0.6, with all corrected item-total correlations above 0.3 However, the Cronbach’s Alpha if Item Deleted for the variable “PBC1 - I think Green SM electric motorbike taxi service is user-friendly” is 0.834, which is higher than the overall scale value Consequently, to improve the scale's reliability, the authors decided to exclude the variable PBC1 and conduct a second analysis of Cronbach's Alpha.
Table 11: Cronbach’s Alpha reliability testing second result of “Perceived behavioral control” scale
Table 12: Cronbach’s Alpha reliability testing result of “Environmental concern” scale
Variables Corrected Item-Total Correlation Cronbach's Alpha if Item Deleted
Perceived behavioral control (PBC): Cronbach ’s Alpha = 0.836
Cronbach’s Alpha value of “Environmental concern" scale is 0.905, greater than
Variables Corrected Item-Total Correlation Cronbach’s Alpha if Item Deleted
Environmental concern (EC): Cronbach's Alpha = 0.905
0.6; corrected item-total correlation of all variables are higher than 0.3 Therefore, all variables in this scale are acceptable and will be used in the subsequent factor analysis.
Table 13: Cronbach’s Alpha reliability testing result of “Price value” scale
Variables Corrected Item-Total Correlation Cronbach’s Alpha if Item Deleted
Price value (PV): Cronbach's Alpha = 0.77Ỉ
The Cronbach’s Alpha value for the “Price value” scale is 0.771, exceeding the acceptable threshold of 0.6, and the corrected item-total correlations for all variables are above 0.3 Consequently, all variables in this scale are deemed acceptable and will be utilized in the forthcoming factor analysis.
Table 14: Cronbach’s Alpha reliability testing result of “Usage intention” scale
Variables Corrected Item-Total Correlation Cronbach's Alpha if Item Deleted
The Usage intention of Green SM (UI): Cron bach 's Alpha = 0.901
The Cronbach's Alpha value for the "Usage Intention" scale is 0.901, indicating strong reliability as it exceeds the threshold of 0.6 Additionally, the corrected item-total correlations for all variables are above 0.3, confirming the acceptability of these variables for further factor analysis.
Table 15: Final results of Cronbach’s Alpha after deleted unsupported variables
The testing of five independent factors and one dependent factor demonstrated a reliability coefficient range of 0.771 to 0.905, with all Cronbach's Alpha values exceeding 0.6 This indicates that the scales utilized in this research are statistically significant and reliable.
Following the analysis of the Cronbach’s Alpha reliability coefficient, the scales underwent further evaluation through Exploratory Factor Analysis (EFA) The Cronbach's Alpha findings revealed 24 observed variables, comprising 20 variables categorized into four factors that assess the "Usage Intention of Green SM," along with four additional variables that contribute to the same scale, all of which satisfied reliability standards Consequently, these 24 observed variables were subjected to EFA testing.
Given that the survey sample size is 390, a combined EFA model was conducted for both independent and dependent variables, employing a minimum factor loading threshold of 0.3.
In the exploratory factor analysis (EFA) of 24 observed variables, Principal Component Analysis with Varimax rotation was utilized, yielding a Kaiser-Meyer-Olkin (KMO) index of 0.820, indicating adequate sampling, and a significant Bartlett’s test of sphericity at 0.000 The analysis revealed a total extracted variance of 73.041%, with all Eigenvalues exceeding 1 for six factors Factor loadings confirmed convergent validity, as all variables exceeded the 0.3 threshold; however, six variables were found to load onto additional factors beyond the initially hypothesized ones.
EC5 - I think humans are seriously abusing the environment: 0.514
EC3 - I think the condition of the environment affects my health quality: 0.507 UI3 - If I use a ride-hailing service in the future, I will think of choosing Green
SM electric motorbike taxi service: 0.354
PBC3 - I have time to use Green SM electric motorbike taxi service: 0.374
PBC2 - I have resources to use Green SM electric motorbike taxi service: 0.339 PV2 - Green SM electric motorbike taxi service offers prices that are suitable for the quality of service: 0.316
AT3 - I support the country in introducing more policies to encourage individuals to use Green SM electric motorbike taxi service: 0.301
Therefore, we decided to raise the minimum factor loading threshold to 0.52 (slightly higher than the 0.514 of variable EC5) and conducted another round of EFA for the 24 observed variables.
In the second iteration of the exploratory factor analysis (EFA) involving 24 variables, Principal Component Analysis with Varimax rotation was performed after raising the minimum factor loading threshold to 0.52 The results showed no significant changes, with the KMO index remaining at 0.820, indicating adequate sampling, and Bartlett’s test of sphericity remaining significant at 0.000 The total extracted variance was consistent at 73.041%, well above the 50% threshold, and all Eigenvalues remained greater than 1 As a result, no variables were excluded, and all 24 observed variables will be included in the subsequent correlation and regression analysis of the model The results of the EFA for these variables are detailed in the accompanying table.
Table 16: KMO and Bartlett tests
Kaiser-Meyer-Olkin Measure of Sampling Adequacy 0.820
Bartlett's Test of Sphericity Approx Chi-Square 6902.924 df 276
EC4 I am concerned about 0.894 environmental issues.
EC2 I think we should have the 0.819 responsibility to protect the environment.
ECI I think environmental issues are 0.818 becoming increasingly serious in recent years.
EC5 I think humans are seriously abusing the environment.
EC3 I think the condition of the 0.641 environment affects my health quality.
UI2 If the Green SM electric motorbike taxi service is good, 1 would recommend friends around me to use.
UI4 I hope Green SM electric motorbike taxi service will be more popular in the future.
UI1 I will use Green SM electric motorbike taxi service in the future.
UI3 If I use a ride-hailing service in the future, I will think of choosing
Green SM electric motorbike taxi service.
SN2 Most people who are important to me think I should use Green SM electric motorbike taxi service in the near future.
SN1 If people around me use Green
SM electric motorbike taxi service, this will prompt me to buy.
SN4 If I use Green SM electric motorbike taxi service, it would be consistent with the trend of social development.
SN3 News media propaganda will prompt me to use Green SM electric motorbike taxi service.
PBC4 Whether or not I use Green SM electric motorbike taxi service is completely up to me.
PBC5 I have sufficient knowledge to 0.841 use Green SM electric motorbike taxi service.
PBC3 i have time to use Green SM 0.691 electric motorbike taxi service.
PBC2 I have resources to use Green
SM electric motorbike taxi service.
PV3 Green SM electric motorbike taxi service generates more value than the cost incurred
PV1 Green SM electric motorbike taxi 0.776 service is reasonably priced.
PV4 Green SM electric motorbike taxi 0.767 service has the suitable price for my needs.
PV2 Green SM electric motorbike taxi service offers prices that are suitable for the quality of service.
AT3 I support the country in introducing more policies to encourage individuals to use Green
SM electric motorbike taxi service.
AT2 I think using Green SM electric motorbike taxi service is a good choice.
AT4 1 think Green SM electric motorbike taxi service satisfies my demand for using a ride-hailing service.
Identifying the correlation between factors and dependent variables
Following the completion of the EFA test, our team conducted a Pearson correlation analysis to examine the relationship between independent and dependent variables, while also addressing potential multicollinearity issues The significance value (Sig.) obtained from this analysis provides insights into the strength and direction of these correlations.
A significant correlation exists between the factors and the dependent variable when the significance level is less than 0.05 and the Pearson Correlation coefficient is greater than 0 Additionally, if the significance value is below 0.05 alongside a high Pearson Correlation, multicollinearity may be present Thus, we conduct Pearson analysis to thoroughly examine each case The results of the Pearson correlation analysis are presented in the table below.
Table 18: Pearson correlation results table
4.3.1 Correlation between Subjective norm and Usage intention
The analysis revealed a statistically significant Pearson correlation (r = 0.453, Sig = 0.000) between the independent variable, Subjective Norm, and the dependent variable, Usage Intention, indicating a positive relationship.
4.3.2 Correlation between Attitude and Usage intention
The Pearson analysis revealed a correlation between the independent variable Altitude and the dependent variable Usage intention, with a Pearson Correlation coefficient of r = 0.440 and a statistically significant p-value of Sig = 0.000 This indicates that as altitude increases, the intention to use also tends to rise, confirming a positive relationship between the two variables.
4.3.3 Correlation between Perceived behavioral control and Usage intention
and a significance level of Sig = 0.000 This indicates a positive relationship, as evidenced by r > 0 and Sig < 0.05.
4.3.4 Correlation between Environmental concern and Usage intention
and a significance level of Sig = 0.000 This indicates that as environmental concern increases, the intention to use the related product also rises, confirming the relationship is statistically significant with r > 0 and Sig < 0.05.
4.3.5 Correlation between Price value and Usage intention
and a significance level of Sig = 0.000 This indicates that as price value increases, the intention to use the product also rises, confirming the relationship between these two variables.
So all 05 independent variables mentioned above are correlated with the dependent variable Usage intention.
Regression Analysis
Sid Error of the Estimate
1 ,769a 0.591 0.585 0.80626 2.208 a Predictors: (Constant), Attitude (AT), Subjective norm (SN), Perceived behavioral control (PBC), Environmental concern (EC), Price value (PV) b Dependent Variable: Usage intention (UI)
The model summary table reveals that the R square value is 0.591, while the adjusted R square stands at 0.585 This adjusted R square indicates that the independent variables in the regression analysis account for 58.5% of the variation in the dependent variable, leaving 41.5% attributed to out-of-model variables and random errors The emphasis on the adjusted R square is crucial, as it provides a more accurate reflection of the model's fit compared to the standard R square.
The Durbin-Watson value of 2.208 indicates that the assumption of first-order series autocorrelation is not violated, as it falls within the acceptable range of 1.5 to 2.5 (Yahua Qiao, 2011).
To test the relationships in the research model between the dependent variable and independent variables, we set the hypothesis:
Hl : HO is not true
Table 20: Analysis of variance (ANOVA)
Model Sum of Squares df Mean Square F Sig.
Total 609.872 389 a Dependent Variable: Usage intention (UI) b Predictors: (Constant), Attitude (AT), Subjective norm (SN), Perceived behavioral control (PBC), Environmental concern (EC), Price value (PV)
The ANOVA table indicates that the significance value of the F-test is 0.00, which is less than the threshold of 0.05 This confirms that the regression model is statistically significant, suggesting that there is at least one causal variable influencing the dependent variable with 95% confidence.
4.4.3 Testing on the partial effects of the independent variables on the dependent variable
To examine which independent variables have statistically significant effects on the outcome variable, we use the t-test.
Standardize d Coefficients t Sig Multicollinearit y Statistics
0.048 0.035 0.049 1.359 0.175 0.829 1.206 a Dependent Variable: Usage intention (UI)
The Coefficients table indicates that the Variance Inflation Factor (VIF) is at a maximum of 1.356, which is below the threshold of 2 This suggests that the independent variables are not correlated, confirming that multicollinearity is not an issue.
The t-test analysis reveals that four independent variables—Attitude (AT), Subjective Norm (SN), Perceived Behavioral Control (PBC), and Environmental Concern (EC)—are statistically significant at the 5% significance level (Sig < 0.05) In contrast, the variable Price Value (PV) does not show statistical significance (Sig > 0.05).
All 04 statistically significant variables Attitude (AT), Subjective norm (SN), Perceived behavioral control (PBC), and Environmental concern (EC) have a positive effect on the dependent variable (regression coefficient 13) The individual effects of these 04 variables align with the initial assumptions of the group Based on these results, the hypothesis testing yields the following summary:
Table 22: Summary Table of Hypothesis Testing
Hl: Attitude has a positive effect (+) on the intention to use Green SM.
H2: Subjective norm has a positive effect (+) on the intention to use Green SM.
H3: Perceived behavioral control has a positive effect (+) on the intention to use Green SM.
H4: Environmental concern has a positive effect (+) on the intention to use Green SM.
H5: Price value has a positive effect (+) on the intention to use Green SM.
So after eliminating the variable that does not affect the dependent variable, the regression equation of the group takes the following form:
UI = 0.31 OAT + 0.180SN + 0.250PBC + 0.374EC
- UI: The intention to use Green SM electric taxi service of students in Ho Chi Minh City
Usage intention \ of Green SM )
4.4.4 Testing the normal distribution of residuals
Testing the normality of residuals in regression is essential for validating statistical inferences and ensuring the reliability of the model Evaluating the distribution of residuals is vital for upholding the assumptions required for hypothesis testing, estimating sample means, and enhancing the accuracy of linear regression analyses To visualize normality, we will utilize Histogram charts, Normal P-P Plots, and Normal Q-Q Plots, along with data tables from the Kolmogorov-Smirnov test and skewness measurements to determine if the index meets the necessary criteria.
Figure 8: Regression Standardized Residual Histogram chart
The histogram chart analysis shows a mean value close to zero at -5.92E-16 and a standard deviation of approximately 1 at 0.994 The model displays a bell-shaped curve with distinct tails, closely resembling a standard distribution This pattern confirms the normality of residuals, indicating that the distribution of residuals adheres to the assumptions of normal distribution.
Figure 9: Normal P-P Plot of Regression Standardized Residual
Normal P-P Plot of Regression standardized Residual
The Normal P-P Plot demonstrates that the residuals are randomly distributed around the diagonal line, closely following it, which supports the hypothesis of a normal distribution This indicates that the collected residuals do not violate the assumptions of normality.
Figure 10: Normal Q-Q Plot of Regression standardized Residual
The Normal Q-Q Plot, like the P-P Plot, indicates that the residuals are randomly scattered around the line and closely adhere to it This pattern suggests that the data behaves as anticipated, confirming that the assumptions of normality are not violated.
To assess normality in the data, we employed the Kolmogorov-Smirnov test for both unstandardized and standardized residuals The results indicated that the significant values for both types of residuals were 0.102, exceeding the minimum threshold of 0.05 Consequently, we conclude that the data meets the assumptions of normality.
The skewness value of -0.735 indicates that the distribution meets the criteria for a normal distribution, as it falls within the acceptable range of -1 to +1 Consequently, this confirms that the assumption of normality for the residuals is upheld.
4.4.5 Testing the assumption of a linear relationship
The research team utilized a scatter plot to examine the linear relationship between the independent and dependent variables, as illustrated in Figure 11 This scatter plot displays the distribution of standardized residual values against predicted values A concentration of data points near the y-axis at 0 indicates that the model's assumption of a linear relationship remains intact.
The data points illustrated in Figure 11 are predominantly clustered at the intersection of the y-axis, demonstrating a clear linear relationship within the model and confirming that the assumption of linearity is met.
Testing One-Way ANOVA
4.5.1 Conduct hypothesis testing on the overall difference in “Intention of using Green SM electric taxi service” between male and female students
A study was conducted to examine the differences in intention to use Green SM electric vehicle services between male and female students The researchers employed Tests of Homogeneity of Variances, which yielded significant results.
Table 25: Tests of Homogeneity of Variances
Levene Statistic dfl df2 Sig.
Based on Median and with adjusted df 621 2 385.343 538
The Sig value of Levene's test = 0.302 (greater than 0.05), indicating that the variances between male and female genders are not different.
Then, we will use the F-test result in the One-Way ANOVA table:
Table 26: One-Way ANOVA Table
Sum of Squares df Mean Square F Sig.
Gender N Mean Std Deviation Std Error Mean
The F-test results show a significance level of 0.556, which is greater than the 0.05 threshold This leads us to accept the null hypothesis (HO), suggesting that there is no significant difference in the means between gender groups Consequently, it can be concluded that there is no variation in the intention to use electric taxi services across different genders.
Due to the lack of difference, we do not estimate separately for males or females but collectively for the entire sample.
4.5.2 Conduct hypothesis testing on the overall difference in “Intention of using Green SM electric taxi service” between first-year, second-year, third-year, fourth year students, and students in other years.
The study aimed to evaluate the differences in the intention to use Green SM electric vehicle services among students in their first to fourth years and those in other academic years To achieve this, the researchers employed the Tests of Homogeneity of Variances, leading to significant findings.
Table 28: Tests of Homogeneity of Variances
Levene Statistic dfl df2 Sig.
Based on Median and with adjusted df 5.374 4 288.573 000
The Sig value of Levene's test = 0.000 (less than 0.05), indicating that the variances between year groups are different.
Then, we will use the Welch-test result in the Robust Tests.
Table 29: Robust Tests of Equality of Means
The Welch test reveals a significance level of 0.036, which is below the 0.05 threshold This leads to the rejection of the null hypothesis (H0), suggesting a notable difference in the means across various year groups Consequently, it can be concluded that there is a variation in the intention to use electric taxi services among different year groups.
4.5.3 Conduct hypothesis testing on the overall difference in “Intention of using Green SM electric taxi service” between usage frequency groups
The study aimed to examine the differences in the intention to use Green SM electric vehicle services among students across various academic years, including first-year through fourth-year students and those in other years To achieve this, the researchers employed Tests of Homogeneity of Variances, yielding significant results that highlight the varying intentions based on students' year of study.
Table 31: Tests of Homogeneity of Variances
Levene Statistic dfl df2 Sig.
Based on Median and with adjusted df 561 3 360.439 641
The Sig value of Levene’s test = 0.313 (greater than 0.05), indicating that the variances between usage frequency groups are not different.
Then, we will use the F-test result in the One-Way ANOVA table.
Table 32: One-Way ANOVA Table
Sum of Squares df Mean Square F Sig.
The F-test revealed a significance level of 0.877, exceeding the 0.05 threshold Consequently, we accept the null hypothesis, which indicates that there is no significant difference in the means of the usage frequency groups This suggests that the intention to use the electric taxi service remains consistent across different usage frequency groups.
Due to the lack of difference, we do not estimate separately but collectively for the entire sample.