The study identifies six factors, including price, ease of use, customer trust, website quality, security, and delivery.. The fust four dimensions are classified as core servrce scales,
Trang 1THAI DONG TAN1*, LY THI THU THAO1, NGUYEN THI PHUONG GIANG1
‘Khoa Thương mại - Du lịch, Trường Đại học Công nghiệp Thành pho Hồ Chi Minh,
*20102301.tan@jSttident iuh edu.vn
Abstracts The outbreak of Covid-19 has led to a significant shift ill consumer behavior, with more consumers choosing to shop online rather tlian in physical stores due to the convenience and time-saving benefits However, maintaining and improving customer satisfaction is a key cliallenge for e-commerce platforms The purpose of this article is to investigate the factors that influence the satisfaction of students
at Industrial University of Ho Chi Minh City (IUH) when making purchases on e-conunerce platforms The study identifies six factors, including price, ease of use, customer trust, website quality, security, and delivery The survey data was collected from 378 IUH students using Google Forms, and the data was analyzed using SmartPLS 4 and SPSS software The results indicate that customer trust and website quality are the two most significant factors that influence customer satisfaction Based on the analysis, the study proposes several management implications that e-commerce platforms can adopt to improve customer satisfaction when purchasing goods online
Keywords. Satisfaction, e-cornmerce platform, online purchase
Tóm tăt. Sụ bùng phát của Covid-19 đã dân đến sụ thay đôi đáng kê trong hành vi của nguôi tiêu dùng, klú nhiều người tiêu dùng lựa chọn mua sắm trục tuyến hon là tại các cửa hàng thực tế do sụ tiện lợi và tiết kiệm thời gian Tuy nhiên, việc duy trì và cải thiện sự liài lòng của klrách hàng là một thách thúc clúnh đối với các nền tảng thương mại điện tử Mục đích của bài viết này là khảo sát các yếu tố ảnh hưởng đến sự liài lòng của sinh viên Trường Đại học Công nglúệp Thành phố Hồ Clú Minh (IUH) klú mua liàng trên nền tảng thương mại điện tử Nglúêrr cứu xác định sáu yếu tố, bao gồm giá cả, dễ sử dụng, niềm tin của khách Iràng, chất lượng trang web, bảo mật và khâu giao liàng Dữ liệu khảo sát được thu thập từ 378 sinh viêrr IUH bằng Google Biểu mẫu vả dữ liệu được pliân tích bằng phần mềm SmartPLS 4
và SPSS Kết quả clrỉ ra rằng niềm tin của khách hàng và chất lượng trang web là Irai yếu tố quan trọng nhất ảnh hưởng đến str liar lòng của klrách liàng Trên cơ sở phân tích nglúên cứu đề xuất một số liàrn ý quản trị mà các sàn thương mại điện tử có thể áp dụng để nâng cao mức độ liài lòng của klrách liàng klú mua hàng trực tuyến
Từ khóa.Sự hài lòng, sàn thương mại điện tử, mua liàng trực tuyến
Currently, the internet develops very fast and is used by many users The number of people participating
in the internet though various apps and platforms has increased much In January 2021, about 68 million people in Vietnam used the internet [1], and as of September 2022, the number of internet users lias increased to more than 72 million, accounting for 73% of the population and an increase of about 4.9% year-on-year [2], All activities of buying and selling and exclranging valuable information on an online
Trang 2platform are considered e-cormnerce Initially, consumers were reluctant to shop online for fear' that their' personal and bank-related information might be exposed to the outside on online shopping forums Moreover, they are not sure whether the purchased product satisfies the desire However, satisfaction with online purchases Iras increased significantly Many consumers believe that tills is the most cost-effective and effective method of shopping The sustainability between customers and e-businesses depends on the quality of service during the purchase process The study of consumer satisfaction when buying online is
an important part of economic research To achieve a successful business strategy, a deep understanding
of customer satisfaction is key Customer satisfaction, in tills case, students, is a crucial factor in building trust and authenticity for a brand Elevating student satisfaction with the online shopping experience can generate a positive encounter for them Tills encompasses providing a user-friendly interface, flexible payment options, excellent customer service, and ensuring timely and quality delivery of products as expected The e-commerce market is competitive, and enhancing student satisfaction can grant you a competitive edge By delivering a superior shopping experience compared to competitors, you can attract
a larger customer base Customer satisfaction, in tills case, students, can offer valuable feedback You can utilize tills feedback to enhance various aspects of your services and products, thereby improving overall quality and creating a better shopping environment Therefore, this is the reason for US to study the topic
“Factors affecting the satisfaction of students at Industrial University of Ho Chi Minh City when purchasing online on e-commerce platforms” The aim of the study was to learn about IUH students' satisfaction when buying online on e-commerce platforms The study will build an appropriate model based on the premise and data analysis results, thereby giving management implications to improve satisfaction when buying on e-cornmerce platforms for consumers
2 THEORETICAL BASIS
2.1 Concept
On September 25, 2021, Decree 85/2021/NĐ-CP was issued The concept of e-commerce platforms is defined in accordance with current legal documents, wliich state that: “An e-commerce trading floor is a website that enables individuals, organizations, and merchants who do not own a website to conduct part
or all of the process of buying and selfing goods and services on it However, tills decree specifically excludes online securities trading websites from the definition of e-cornmerce trading floors.” [3]
“Customer satisfaction is the perception of a good or service that meets or exceeds their' expectations in the process of consuming that good or service” (Kotler, 2001) [4], “Satisfaction is the perception and judgment of customers that the quality of products or services is suitable for their consumption purposes” (Srivastava et al., 2013) [5],
2.2 Theoretical model
Theory of Reasoned Action - TRA
TRA posits that an individual’s behavioral intention determines their actual beliavior Behavioral intention
is determined by the individual’s attitude towards the behavior and subjective norm related to performing the behavior Attitude towards behavior is expressed through the individual’s belief about the likelihood
of the behavior producing desired outcomes, which can be evaluated as favorable or unfavorable Subjective norm reflects the individual’s perception of the correlation between a specific behavior and what reference groups tliink about tills beliavior
Theory of Planned Behavior - TPB
TPB is an extension of the TRA TPB can better explain individual behavior in specific contexts where the individual cannot fully control their' behavior, specific behavior perception is determined by both past experience and anticipated barriers
Technology Acceptance Model - TAM
Among theories in information systems, the Technology Acceptance Model (TAM) developed in 1989 by Davis is identified as one of the most widely used analytical frameworks to explain consumer behavioral intention when using a system TAM explains that a consumer’s behavioral intention to use a system is influenced by perceived ease of use and perceived usefulness
SERVQUAL Model
© 2023 Tnròng Đại học Công nghiệp Thành phố Hồ Ơ1Í Minh 33
Trang 3SERVQUAL is a tool that assesses service quality by comparing the gap between customer expectations and perceptions of a service provided, bl order to use SERVQUAL, participants are requited to provide feedback on both their expectations and perceptions of the service However, since SERVQUAL measures customer perceptions rather than the actual quality of service delivered, it primarily captures attitudes related to satisfaction rather than providing an objective measure of service quality Tills makes
it different from other tools used for measuring servrce quality (Parasuraman et al., 1988) [6]
E-SERVQUAL Model
Zeithaml et al (2005) [7] developed the e-SERVQUAL e-Service Quality Measurement model to investigate how customers evaluate e-service quality The development of this model involved three stages, including the identification of focus groups, data collection, and analysis of empirical data The e- SERVQUAL model comprises of seven dimensions, namely efficiency, reliability, responsiveness, privacy, compensation, contact, and assurance The fust four dimensions are classified as core servrce scales, while the remaining three dimensions are considered as recovery scales since they become prominent only when online customers face any issues or have queries
Through exploring many different factors, tills study focuses on 6 main factors as follows:
3.1 Price
Price is defined as the amount customers pay for the benefits they receive from a product or service (Lovelock & Wirtz, 2007) [8], According to Oliver (1997), customers perceive the equity value of a product or service and judge their satisfaction or dissatisfaction based on that perception [9], As sellers can easily change prices online (Karman & Kopalle, 2001) [10], the price of a product purclrased online is usually lower (Forsythe et al., 2006) [11], Therefore, online shoppers tend to be more satisfied with a lower price (Changsu Kim, Weihong Zhao, Kyung Hoon Yang, 2008) [12], Additionally, research by Ralfi, Ghani, and Muhamad (2017) shows that buying at the right price positively affects customers' confidence when purchasing online [13], Based on these findings, the following hypothesis is proposed:
Hl: Price positively affects customer satisfaction
H3: Price positively effects customer confidence
3.2 Customer tr ust
Srderslmiukh et al (2002) argue that customer trust is crucial for service providers to deliver on their promises and be perceived as reliable [14], According to Mukherjee and Nath (2003), trust plays a vital role as a foundation for cooperation between sellers and buyers [15], Customer trust and satisfaction are strongly connected, with satisfaction even being considered a component of trust (Sahadev & Purani, 2008) [16], rust in service marketing places significant emphasis on individual attitudes, beliefs of customers, and the reliability of services received (Garbarino & Lee, 2003) [17], Therefore, the hypothesis is proposed as follows:
H2: Customer trust positively affects customer satisfaction
3.3 Security
bl tills study, security is defined as "the abfiity to protect a customer's data during an online purchase." Consumers are highly concerned about the processing, storage, and usage of their personal data when making purchases, and they require assurance of secure transactions (Libaque-Saenz et al., 2021) [18], Customer information security is thus a crucial factor in ensuring customer satisfaction (Valdez-Juarez et al., 2021) [19], Retailers must ensure strict control over the ordering and online payment processes To assuage consumers' apprehensions regarding the security of online transactions, online retailers employ a range of security measures, including encryption, safeguarding, verification, and authentication (Chellappa and Pavlou, 2002) These mechanisms safeguard consumer data from unauthorized viewmg or alteration and guarantee that only relevant entities (such as vendors, credit card authorizes like Visa, and banks) possess access to consumer information Consumer perceptions of these security measures significantly bolster then trust in online transactions (Chellappa and Pavlou, 2002) Online retailers' privacy and security poheies play a critical role in enhancing customer trust, and they act as a signal that
Trang 4influences customer trust (Martin and Carnarero, 2009) [20], Therefore, the following hypothesis is proposed:
H4: Security positively affects customer satisfaction
H5: Security positively affects customer trust
3.4 Ease of use
hl then' 2016 study, Tandon et al provided a definition for ease of use, which refers to the extent to which individuals are able to comprehend, learn, and effectively utilize specific technologies or systems [21], When creating an online shopping website, it is essential for businesses to prioritize ease of navigation, education, awareness, and support In the development of e-commerce, ease of use plays a critical role, with attributes like product search and information liaving the potential to generate favorable ratings in online retail Customers are more likely to be satisfied with their purchases if they can easily locate the items they are looking for and obtain relevant information about them Research conducted by Tu et al in
2012 revealed that perceived ease of use has a positive impact on customer satisfaction [22], Similarly, Aghdaie et al found that it has a positive impact on trust in a 2011 study [23], Koufaris et al also discovered a positive correlation between ease of use and trust in then 2002 research [24], Furthermore, stirdies in the e-conunerce industry by Chinornona (2013) [25], Wilson (2019) [26], Keni (2020) have all demonstrated a favorable impact of ease of use on trust [27]
H6: Ease of use positively effects customer satisfaction
H7: Ease of use positively effects customer trust
3.5 Website quality
A company's website plays a crucial role in determining service quality, which in turn affects customers' buying experience and their ability to find relevant information easily (Kim & stoel, 2004) [28], An informative and well-organized website can effectively persuade customers to make a purchase (Wolfinbarger & Gilly, 2001) [29], as it saves them time searching for product infonnation and eliminates misinformation (Luo, Ba & Zhang, 2012) [30], Thus, designing a website tliat is credible and trustworthy
in the eyes of customers is crucial for online service providers (Guo et al., 2012) [31], Therefore, a company's website is considered a significant factor in determining customer satisfaction with online services
H8: Website quality positively affects customer satisfaction
3.6 Delivery
hl today's liiglily competitive business environment, providing door-to-door dehvery services is essential
to meet customers' expectations and ensure their satisfaction when making online purchases Customers place special emphasis on delivery time and method, including issues such as delivery delays, shippers' attitudes, wrong deliveries, improperly packaged goods, and damages (Forsythe et al., 2006) [32], OlU' company lias received several complaints regarding product delivery, with most focusing on delayed or incorrect deliveries and shipping costs (Vu Huy Thong & Tran Mai Trang, 2013) [33], It can be concluded that the delivery process in e-commerce is a crucial step that suppliers and store organizations must carefully manage to ensure successfid delivery of products to customers
H9: Delivery positively affects customer satisfaction
© 2023 Tnròng Đại học Công nghiệp Thành phố Hồ Ơ1Í Minh 35
Trang 5Figure 1: Proposed Research Model
The research process utilized both qualitative and quantitative methods Qualitative methods were used to collect data from various sources and involved group discussions to identify the factors that influence IUH students' satisfaction when making purchases on e-commerce platforms Following the completion
of the qualitative research, the findings were used to develop a draft model To further the study, quantitative research methods were employed by conducting an extensive survey to a broader audience A quantitative questionnaire was designed using a 5-point Likert scale to gather opinions from the majority
of IUH students, and statistical analysis was carried out to measure the significance of the factors identified through the qualitative research A total of 451 samples were collected, and after filtering and cleaning, 378 samples were considered satisfactory for analysis The data was analyzed using two primary statistical software, SmartPLS 4 and SPSS, and the analysis techniques included descriptive statistics, Cronbach's Alpha reliability testing, exploratory factor analysis (EFA), and structural equation modeling (SEM) validation
The author adhered to the minimum sample size rule of 5 times the number of observed variables when conducting the EFA, which was 140 samples since this study had 28 observed variables (Hah' et al., 2014) [34], The sample size of 378 was deemed sufficient and appropriate for analysis in tills study
5.1 Descriptionof the survey sample
Table 1.Descriptive statistics of tile survey sample
Coruse
Income
Trang 65.2 Exploratory factor analysis (EFA)
The results of the Exploratory Factor Analysis (EFA) using the Principal Component's extraction method and Varimax rotation indicate the following:
Out of the 28 observed variables measured, 6 underlying factors influencing satisfaction were extracted with a Kaiser-Meyer-Olkin (KMO) measure of 0.913 and Sig = 0.000 The extracted variance accounts for 69.182%, with an Eigenvalue of 1.293 Additionally, all variables exhibit factor loadings that meet the requirement (> 0.5)
Furthermore, 3 observed variables from the Satisfaction scale were extracted into the same factor, with a KMO measure of 0.726 and Sig = 0.000 The extracted variance accounts for 76.015%, with an Eigenvalue of 2.280
All variables also meet the requir ed factor loading criterion (>0.5) Thus, the results of the Exploratory Factor Analysis (EFA) for both independent and dependent variables are deemed appropriate
Table 2 Results of EFA analysis for independent variable
Rotated Component matrix*
Component
5.3 Testing the measurement model
To assess the reliability of the measurement scale, the study used Composite Reliability (CR), Average Variance Extracted (AVE), and outer loading coefficient According to the results presented in Table 2,
© 2023 Tnròng Đại học Công nghiệp Thành phố Hồ Ơ1Í Minh 37
Trang 7all variables have a Cronbach’s Alpha coefficient greater than 0.7, with the WQ variable having the smallest coefficient of 0.187 and the EU variable having the highest coefficient of 0.891 The CR coefficients for all variables are greater than 0.7, ranging from 0.891 to 0.920 [35], The AVE coefficients are all greater tlian 0.5, with the smallest being the sc variable at 0.646 and the largest being the ST variable at 0.760 [36], According to Table 3, the outer factor loading coefficients of all observed variables are greater than 0.7, ranging from 0.722 to 0.891 [37], Therefore, the measurement model is acceptable, with high reliability and convergence when Cronbach’s Alpha and CR coefficients are both greater tlian 0.7 and AVE is greater than 0 5
Table 3: Results on tire reliability of the measurement scale and convergence values
Security (SC) 0.863 0.901 0.646 Website Quality (WQ) 0.817 0.891 0.732 Ease of use (EƯ) 0.891 0.920 0.697 Price (PR) 0.841 0.893 0.677 Delivery (DL) 0.856 0.893 0.676 Customer trust (CT) 0.838 0.892 0.673 Satisfaction (ST) 0.842 0.905 0.760
Table 4: Results of outer loading factors
Trang 8The research team conducted tests on the measurement model to verify its accuracy, specifically by assessing the discriminant validity of the factors To aclúeve this, two criteria, namely Fomel-Lacker and HTMT, were utilized Based on the Fomel-Lacker criterion, discriminant validity is exlúbited in table 4, where the square root of the average variance extracted (AVE) for each factor is greater than the correlation coefficient of other factors in the same column Tills finding indicates that all factor's possess satisfactory discriminant validity Furthermore, the HTMT criterion also affirms the discriminant validity
of the factors, with coefficient values below 0.85 as displayed in table 5 Therefore, the factors can be concluded to have adequate discriminant validity [38], [39],
Table 5: Fomel standard - Lacker
Table 6: Heterotraite-Monotrait Index (HTMT)
sc
5.4 Structural model verification
According to the results presented in table 6, the VIF coefficients range from 1.707 to 2.368, wlfrch are both below the threshold value of 3 This indicates that there is no occurrence of linear' multi-additive phenomena in the measurement model, and as a result, the hypothesis test results of the study are not affected [40],
© 2023 Truông Đại học Công nghiệp Thành phố Hồ Ơ1Í Minh 39
Trang 9Table 7: VIF coefficient
The results ill table 7 show that the R-square adjuster values of the Satisfaction and Trust models are 0.487 and 0 371, respectively Analyzing the impact of factors on satisfaction shows that factors related to security, website quality, piice, delivery, and ease of use, along with customer trust, explain 48.7% of the variance in satisfaction, with statistical significance at the 5% level The remaining 51.3% is due to system enor and other factors that have not been included in the model
Table 8: R - Square and R - Square adjusted
R - Square R - Squar e adj listed
The results in table 8 show the level of impact of the factor's evaluated through the f3 value, indicating the strongest impact of CT on ST with an f3 value of 0.065, followed by WQ, sc, PR, EU, and DL bl addition, sc and EƯ have a more significant impact on CT, wliile PR lias a lower impact with an f3 value
of 0.023
Table 9: F-squared index
Trang 10Figure 2: PLS SEM results formeasurement models Because the data ill PLS path analysis does not conform to a normal distribution, traditional significance tests of parameters in regression analysis are not applicable for testing the significance of coefficients, hl its place, PLS relies on the non-parametric bootstrapping technique to determine the significance of coefficients (Hair et al., 2014) [34], To determine if the path coefficient significantly deviates from 0, the t-value is computed using bootstrapping In tills study, the non-parametric bootstrapping technique was employed on a sample size of 378 with 5000 iterations to guarantee the precision of the lineal' structural model test
Table 10: Results of the impact levels of relationships Original
sample (O)
Standaid deviation (STDEV)
T statistics (IO/STDEVI) p values Hypothesis Result
© 2023 Tnròng Đại học Công nghiệp Thành phố Hồ Ơ1Í Minh 41