INTRODUCTION
Research background
In today's digital age, information technology plays a crucial role in shaping social life, enabling easy access to the latest innovations Recent advancements, particularly in mobile apps, Artificial Intelligence (AI), and the Internet of Things (IoT), are set to transform the mobile app development industry With over three billion smartphone users in 2020, this number is expected to exceed four billion by 2023, presenting new opportunities for businesses worldwide The surge in mobile app downloads has been remarkable, rising from 140 billion in 2016 to 218 billion in 2020, highlighting the growing demand for mobile applications.
The rise of online business, particularly through mobile interactions, is transforming the way companies engage with customers, boost sales, and enhance brand loyalty Mobile applications, designed for smartphones and tablets, facilitate this connection, providing businesses with effective channels to promote their products and services Information technology streamlines shopping, transactions, and payments, allowing consumers to shop from the comfort of their homes E-commerce has largely supplanted traditional markets, enabling families to enjoy convenient home shopping experiences The growing popularity of food ordering and delivery applications exemplifies this trend, catering to consumers' changing habits in the technology-driven era Customers increasingly favor fast delivery, even for single-item orders, while restaurant owners benefit from expanded reach and profitability through these platforms This shift reflects a modern lifestyle where convenience and speed are paramount, attracting busy individuals willing to pay for quick service.
According to the Adsota Vietnam Digital Advertising Market report (2020), Vietnam boasts 43.7 million smartphone users, accounting for 44.9% of its 97.4 million population, positioning the country among the top 15 global markets for smartphone usage With a population exceeding 90 million, Vietnam is on the brink of a digital explosion, driven by the influence of modern urban lifestyles and the prevalence of technology, smartphones, and e-wallets The significant presence of Millennials and Gen Z has transformed consumption and eating habits, leaning towards convenient, door-to-door solutions Busy young professionals and office workers increasingly prefer online food ordering, highlighting a growing trend in major cities like Hanoi and Ho Chi Minh City.
Da Nang, where many online and high-tech companies are concentrated
The ongoing Covid-19 pandemic has led many countries to implement stricter social distancing measures, resulting in economic stagnation and recession However, advancements in technology have facilitated remote work, ensuring productivity and effective communication from home This shift has allowed businesses to operate online, even as traditional storefronts close Consequently, there has been a notable increase in online food orders as consumers prefer delivery over dining out, prompting many restaurants to pivot to online services to meet this demand and mitigate revenue losses As a result, the food delivery model has surged in popularity, reflecting a significant shift in consumer behavior during the pandemic.
Many restaurants, including popular brands like King BBQ, Gogi House, and Pizza 4P's, are adopting the "living with the flood" solution to navigate challenges The rise of online business has become a significant opportunity, allowing restaurants to not only increase their customer base but also tap into new markets While the Covid-19 pandemic has had negative effects, it has also led to a lasting shift in consumer eating habits, with a growing preference for online food ordering even as the situation stabilizes.
A recent Nielsen Vietnam survey (2020) revealed that 62% of Vietnamese consumers preferred home food purchases during the Covid-19 pandemic, with 19,000 restaurants joining the food delivery network Euromonitor International projects that Vietnam's online food delivery market will reach approximately $38 million by 2020, growing at an average rate of 11% over the next five years Kantar TNS research indicates an even higher revenue growth rate of 28.5% per year, potentially reaching $449 million by 2023 Despite its modest current scale, the online food delivery market in Vietnam has significant growth potential, especially in urban and satellite areas, offering restaurants opportunities to diversify income streams and reduce reliance on traditional dining models amid ongoing pandemic challenges Vietnam is rapidly evolving into a pioneering market in this sector.
In 2020, online meal ordering surged, with one in ten meals being ordered online, primarily for lunch and snacks Over 75% of users in Ho Chi Minh City and Hanoi place online orders at least weekly, with nearly 30% ordering 2-3 times a week and about 5-6% ordering more than ten times This trend highlights the significant need for restaurants and eateries to embrace online platforms, leveraging the substantial benefits of ordering apps to adapt to consumer behaviors and sustain operations during crises Apps like Gojek's GoFood utilize advanced technology, including AI and automation, to better understand customer preferences and create tailored offers, enabling restaurants to reach a diverse customer base without geographical limitations and facilitating rapid order fulfillment on a large scale.
A 2020 market survey by Kantar revealed that 43% of residents in Ho Chi Minh City and 34% in Hanoi order food online at least once a week Due to pandemic concerns, consumers prefer home delivery over take-away, with home delivery accounting for approximately 30% of Vietnam's F&B market share and showing significant growth The rise in users of food ordering applications has intensified competition among them, offering convenience for urban dwellers during social distancing measures These applications enable people to enjoy their favorite meals without leaving home, allowing restaurants to continue operations amid challenging circumstances.
Research Motivation
Mobile applications are poised to emerge as a crucial sales channel for retailers, highlighting the need for prompt analysis of customer perceptions (Magrath & McCormick, 2013) In the restaurant sector, mobile food ordering apps (MFOAs) are increasingly recognized as innovative tools for engaging customers and delivering exceptional service.
In recent years, the Vietnamese market has witnessed a significant surge in Multi-Functional Online Applications (MFOAs), despite the fact that research on these platforms remains in its infancy While online food ordering apps have garnered considerable attention, academic studies examining the intricacies of MFOAs are still limited As these applications have only recently gained popularity in Vietnam, it is essential to explore the factors that influence customer perceptions, intentions, and behaviors regarding their use.
Consumers have diverse interests and expectations regarding food quality and service when using mobile food ordering apps Research indicates that customer satisfaction levels with these applications can vary significantly (Cho et al., 2019) However, there is a lack of studies identifying the key attributes of mobile food ordering apps and examining how these quality factors influence user perceptions and their intentions to reuse or recommend these apps in Vietnam.
The rise of food delivery services offers significant advantages for both restaurant owners and customers This efficient channel minimizes resource consumption while allowing businesses to reach a broader audience, leading to increased revenue For consumers, ordering food via apps is a quick, cost-effective solution that provides a wide range of options to choose from.
This study investigates customer satisfaction and the ongoing intention to reuse multi-functional online applications (MFOAs) in Vietnam, where these platforms are widely accepted It aims to identify key factors that may either facilitate or obstruct the effective implementation of MFOAs.
Research Objective and Research Questions
Mobile food ordering apps (MFOAs) are increasingly recognized in the catering industry as innovative tools that can enhance customer engagement and service quality Despite their potential, uncertainties persist regarding the impact of MFOA implementation on user satisfaction and the likelihood of repeat usage This study aims to clarify the factors influencing customer satisfaction with MFOAs, identify key predictors of satisfaction and continued usage in Vietnam, and propose actionable strategies for organizations to boost user satisfaction with their apps.
Question 1: What are the factors influencing customer satisfaction of MFOAs in Vietnam?
Question 2: Does Various Food Choices positively influence customer satisfaction of MFOAs in Vietnam?
Question 3: Does Customer Satisfaction have postive impact on customers’ continued intention to use MFOAs in VN?
Significant of the research
This study aims to enhance the understanding of Mobile Food Ordering Applications (MFOAs) in Vietnam by identifying and experimentally examining the key factors that affect consumer satisfaction and their intention to reuse these applications An integrated model is proposed, drawing from the Catering App Success Model by Y S Wang et al (2019), which itself is based on the E-commerce System Success Model (Wang & Tang, 2003), while also considering the unique characteristics of MFOAs in Vietnam, particularly the diverse food choices available Additionally, the research introduces new measurement scales tailored to the MFOA context in Vietnam to assess the impact of Information Quality, System Quality, and Service Quality on customer satisfaction.
This study aims to assist MFOAs providers in Vietnam by identifying the key factors that influence user satisfaction and their intention to continue using MFOAs By understanding these elements, providers can enhance their MFOA offerings and services, ultimately attracting more users and strengthening customer loyalty.
LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT
Literature review
2.1.1 Mobile Food Ordering Apps (MFOAs)
Mobile applications are increasingly popular among businesses for engaging potential customers Designed specifically for handheld devices like smartphones and tablets, these apps represent one of the fastest-growing sectors in the enterprise software market They serve as a key channel for customers to explore updated products and connect with retailers without needing to be physically present.
Mobile app developers and merchants can leverage the inherent qualities and techniques of mobile applications to influence users' emotions, thereby enhancing online shopping behavior (Oh et al., 2010) The term "app," short for "application," encompasses all software that operates on various operating systems.
MFOAs, or mobile food ordering applications, enable customers to conveniently order food using mobile devices and dedicated apps (Okumus & Bilgihan, 2014; Alalwan, 2020) These platforms serve as effective channels for restaurants to engage and inform busy customers, enhancing their overall dining experience (Okumus & Bilgihan, 2014).
MFOAs offer innovative solutions that help users and businesses tackle challenges like long wait times, traffic issues, miscommunication, delayed deliveries, and customer complaints By utilizing MFOAs, users can conveniently and efficiently access and order meals from various restaurants at their preferred times and locations.
According to Yeo et al (2017), food delivery services can be categorized into two main types: first, restaurants and eateries that manage their own delivery services, such as Pizza Hut, McDonald's, and Domino's Pizza; and second, intermediaries that facilitate food ordering and delivery from various restaurants, including platforms like Food Panda, Room Service, GrubHub, and Eat24hours.com.
Over the past three years, Vietnam's food delivery service industry has experienced significant growth, becoming a dominant force in the market with the involvement of major players Euromonitor's research predicts that the food delivery market in Vietnam will reach a value of $38 million by 2020.
In 2020, Vietnam witnessed a significant surge in food ordering application users, intensifying competition among various platforms These applications provided essential convenience during the complicated pandemic situation, allowing users to order their favorite meals from home while enabling restaurants to maintain operations amid social distancing measures Undoubtedly, food ordering apps have enhanced the daily lives of urban residents, making dining more accessible and convenient.
Vietnam's mobile food delivery service market is highly competitive, dominated by major players like GrabFood, Now, Baemin, GoFood, and Loship A 2020 survey by Rakuten Insight revealed that 66% of respondents preferred GrabFood as their primary mobile food ordering app, with Now at 41%, GoJek at 17%, Baemin at 11%, and Loship at just 3%.
The rise of food delivery services offers significant advantages for both restaurant owners and customers This efficient channel allows food establishments to reach a broader audience while minimizing resource consumption, ultimately boosting revenue For consumers, ordering through apps is a fast, cost-effective option that provides a diverse range of culinary choices.
In Vietnam, the leading mobile food ordering apps include GrabFood, Now, GoJek, Baemin, and Loship, which serve as intermediaries for food delivery from various restaurants (Insight, 2020) This article specifically examines the category of food ordering apps identified by Yeo et al (2017), focusing on the most widely used app in Vietnam.
Initially, Now.vn was the leading player in the delivery service market However, as the market expanded, numerous businesses began to emerge, leading to the rise of various online ordering apps like GrabFood, Gojek, and Baemin.
In the digital era, consumer behavior is evolving, with individuals increasingly seeking products and services that enhance convenience and simplicity in their busy lives Experts highlight that businesses must differentiate themselves to cater to these changing customer needs In the competitive online food delivery market, companies with strong capital and technological backing are engaged in a "burning money" race, compelling them to carefully evaluate and prioritize key factors that will effectively satisfy user demands.
With the rise of delivery businesses, customers now enjoy a wider selection of service providers, allowing them to choose those that offer attractive policies, quality service, and prompt delivery This competitive landscape pressures less efficient companies to improve or risk being eliminated from the market Consequently, delivery services are evolving to not only prioritize quick delivery and food quality but also to focus on the attitude of delivery personnel, ensuring enhanced customer satisfaction.
Numerous studies, including those by Kapoor & Vij (2018) and Wang et al (2019), have focused on the effects of Mobile Food Ordering Applications (MFOA) on customer satisfaction, experience, and conversion rates Wang et al (2019) developed a model rooted in the Information Systems Success Model, drawing from the E-commerce System Success Model by Wang & Tang (2003) and relevant marketing literature, to predict the key outcomes associated with customer usage of mobile catering applications.
The author conducted a literature review to find a suitable theoretical framework for evaluating the value of mobile business applications from the end user's perspective, ultimately identifying the D&M Information Systems Success Model (DeLone & McLean, 2003) as a robust foundation for the study Unlike other models such as the Technology Acceptance Model (TAM), the D&M IS Success Model offers a comprehensive approach to evaluation, supported by extensive empirical research that validates its relationships and provides numerous measures for assessing success dimensions This model's focus on quality dimensions from the end user's viewpoint allows for the comparison of success factors across different applications and technologies, making it a valuable tool for this research.
(2003), ‘‘To assess the performance of a single system, ‘information quality' or ‘system quality' may be the most significant quality component,"
Research Model and Hypothesis development
Information quality is crucial for mobile applications, focusing on the usefulness, understandability, and timeliness of the information provided to users It plays a vital role in the success of information systems, as highlighted by Mckinney et al (2014) DeLone & McLean (2003) emphasize that 'information quality' or 'system quality' is often the most significant component in evaluating the success of an individual system Furthermore, DeLone & McLean (1992) assert that the quality of information directly impacts user satisfaction and utilization, ultimately influencing user behavior.
The integration of technology into businesses significantly enhances both customer experience and operational efficiency, especially in today's digital age where the internet is essential With constant promotions and appealing advertisements on ordering applications, user demand for online spending has surged These applications are designed with intuitive interfaces and visually appealing images, effectively attracting more customers Additionally, collaborations with payment partners and brand owners enable the implementation of promotions that offer substantial benefits to consumers As a result, the food ordering service industry is experiencing remarkable growth.
Therefore, the following hypothese are proposed:
H1 Information quality positively influences customer satisfaction
System quality refers to the ease of use of mobile business applications, encompassing aspects such as performance, interface design, and navigation (Legner et al., 2016) With the rising popularity of smartphones, online payments have become more convenient, particularly among young people and office workers who favor e-wallets for their efficiency in transactions In the Vietnamese market, various e-wallet gateways like Viettelpay, Momo, Zalopay, Airpay, Moca, and Vnpay offer diverse options and promotions to attract customers, influencing their consumption behavior When using catering apps, consumers prioritize an efficient ordering process, which necessitates a user-friendly and stable system (Wang et al., 2019).
In Vietnam, the convenience of food delivery services has transformed the way customers access meals, eliminating the need to brave unfavorable weather or crowded traffic With just a smartphone or computer, users can effortlessly browse menus, place orders, and receive their food at home As restaurants adapt to online selling, they optimize food preparation and packaging to maintain quality during delivery This service primarily caters to tech-savvy young individuals, such as students and office workers, who seek new experiences and have limited free time Consequently, the rapid growth of the food delivery app industry is revolutionizing the purchasing landscape in Vietnam.
Therefore, the following hypothese are proposed:
H2 System Quality positively influences customer satisfaction
Service Quality encompasses the comprehensive support offered by a service provider in relation to mobile business applications, focusing on factors such as responsiveness, empathy, dependability, and the competency of support personnel (Legner et al., 2016) In the context of catering apps, characteristics like problem-solving abilities and personalized attention serve as key indicators of product quality (Y S Wang et al., 2019) Consumers typically use these applications to place food orders, and when issues arise, they seek tailored assistance from online customer care for resolution High-quality service correlates with high-quality products, leading to increased consumer satisfaction and repeat usage of the catering app.
Research by GCOMM (2019) indicates that urban residents in Vietnam frequently use online ordering services, highlighting significant market potential The study identifies five key factors influencing customer choice in food delivery: fast delivery speed (65%), neat and clean packaging (58%), guaranteed food quality (56%), accurate order fulfillment (50%), and a variety of affordable dishes (45%) Among the food delivery options available, GrabFood stands out as the fastest service, with approximately 80% of customers recognizing it as the leading choice in Vietnam.
A survey by the Vietnamese market research firm Q&Me in December 2020, involving 1,046 participants aged 18 to 45 from Hanoi and Ho Chi Minh City, revealed significant insights into the rising popularity of food delivery apps The primary motivation for using these apps was to save travel time, particularly in cities plagued by frequent traffic jams Notably, 55% of respondents cited laziness to go out as a reason, while 43% appreciated the fast delivery times offered by these services.
As the delivery service market has expanded, Now.vn initially dominated but has faced increasing competition from online ordering apps like GrabFood, Gojek, and Baemin This growth has provided customers with more options, allowing them to choose providers based on favorable policies, service quality, and delivery speed Consequently, delivery businesses must adapt by ensuring fast delivery, maintaining food quality, and enhancing the attitude of their delivery staff to boost customer satisfaction For instance, GrabFood allows customers to rate delivery personnel after receiving their orders, addressing issues related to delivery attitudes and ultimately improving service quality and user loyalty.
Therefore, the following hypotheses are proposed:
H.3 Service quality positively influences customer satisfaction
Sales promotions are short-term incentives aimed at encouraging the trial or purchase of specific products or services, often utilizing mobile coupons to offer price reductions (Wang et al., 2019) Research indicates that customers' attitudes towards mobile coupons and their sense of control significantly influence their likelihood of redeeming these offers (Dickinger & Kleijnen, 2008) When catering apps provide more mobile discounts, users are more inclined to purchase these items, enhancing their transaction utility (Wang et al., 2019) To boost customer satisfaction, companies can either improve the actual quality of their services or elevate perceived service quality through effective promotions (Ye-Eun Song et al., 2017).
The rise of smartphone applications is transforming food-selling brands by enabling them to engage customers more effectively, moving beyond traditional business models Even when customers aren't making purchases, they remain influenced by promotional images and E-voucher codes from food delivery apps This technological advancement is a key factor driving the growth of the food delivery market.
The rising popularity of smartphones and the convenience of online payments have led young people and office workers in Vietnam to favor e-wallets for transactions, benefiting both shippers and buyers The Vietnamese market offers a diverse range of e-wallet options, including Viettelpay, Momo, Zalopay, Airpay, Moca, and VNpay, which are actively using promotions to attract customers and shape consumer behavior.
A 2020 survey by Reputa, a Social Listening Platform in Vietnam, revealed that over 29,500 positive user feedback responses were predominantly linked to the use of discounts and promotional services The analysis indicated that "Incentives and promotions" were the primary driver of customer satisfaction, comprising 84% of the feedback.
A December 2020 survey by Vietnamese market research company Q&Me, involving 1,046 participants aged 18 to 45 in Hanoi and Ho Chi Minh City, highlighted key trends in food app usage The findings revealed that promotional vouchers and delivery costs significantly influence user decisions, with 30% of respondents indicating they only order when promotions are available, while 38% actively seek out promotions.
Therefore, the following hypotheses are proposed:
H.4 Perceived Promotions positively influences customer satisfaction
Y S Wang et al (2019) defined price value as monetary sacrifice that considered by customers when it comes to MFOAs Similarly, price value is conceptualized as the ratio between perceived benefits obtained and monetary sacrifices made by consumers (Sweeney
According to Soutar (2001) and Alalwan (2020), price value is linked to the financial implications of adopting new products and systems Customers tend to weigh the benefits of a new system against its financial costs This study adopts the same definition, specifically examining the advantages of purchasing food through Mobile Food Ordering Applications (MFOAs) compared to traditional purchasing methods.
METHODOLOGY
Sample and data collection
This quantitative research study utilized online questionnaires to gather sample data for hypothesis testing and to meet specific research objectives Participants were individuals in Vietnam who have experience using mobile food ordering applications (MFOAs) to order meals.
Before conducting the survey, the author carefully considered the sample size, recognizing its importance for Structural Equation Modelling (SEM) to test the hypotheses effectively Research indicates that a sample size of less than 100 is small, 100 to 200 is medium, and over 200 is large (Kline, 2005) Other studies confirm that a minimum sample size of 100 to 150 individuals is typically sufficient for analysis (Ding et al., 1995; Rezaei, 2015) For this study, a minimum of 100–150 samples was deemed acceptable for evaluating the model using PLS-SEM Furthermore, Delice (2001) highlights that larger sample sizes enhance the accuracy of research findings.
The questionnaire consists of four parts: the first part screens participants to confirm their experience with MFOAs, allowing only those with prior usage to proceed The second part gathers demographic information, including gender, age, location, occupation, monthly income, and attitudes towards MFOA usage frequency and duration The third part employs measurement scales derived from previous studies to assess various factors Finally, the last section poses additional questions to determine if respondents are interested in trying other MFOAs and explores the reasons influencing their choices, which will enhance the study's implications.
The questionnaires were initially translated into Vietnamese and distributed as a bilingual English/Vietnamese survey A group of 11 postgraduate students from Vietnam Japan University reviewed the translations to identify any misunderstandings or ambiguities, ensuring content validity To gather further feedback, the author conducted a pilot test with 74 individuals experienced in MFOAs Based on the feedback from this pilot test, modifications were made to the structure and wording of the survey.
Data collection for the online survey occurred from May 11 to 25, 2021, with 311 individuals participating After removing invalid responses, the final valid sample consisted of 282 participants, meeting the necessary sample size criteria.
Measurement
This study utilizes measurement items adapted from prior research, along with self-developed items created by the author to align with the specific context of the study.
In this study, a 5-point Likert scale was utilized, with ratings ranging from 1 for "strongly disagree" to 5 for "strongly agree." Research by Dawes (2008) suggests that using a five- or seven-point scale typically yields higher mean scores compared to a ten-point scale, facilitating data comparison The decision to implement a five-point scale aligns with findings from Ye-Eun Song et al (2017), which indicate that fewer scale points enhance the ease of conducting surveys.
Table 3.1: Table of measurement items
No Code Measurement Items References
IQ1 The MFOA provides precise information Doll & Torkzadeh
Y S Wang (2008); IQ2 The MFOA provides up-to-date information
IQ3 The MFOA provides useful and relevent information Ye-Eun Song et al
(2017) IQ4 The MFOA provides reliable information
IQ5 The MFOA does not have much unneccessary information Self-developed
SYQ1 The MFOA is user friendly Doll & Torkzadeh
SYQ2 The MFOA is easy to use
SYQ3 The MFOA has high reliability without errors
The MFOA has variety of functions (examples: food ordering, GPS tracking, different kinds of payment, ways of communication with shippers,…) Self-developed
SYQ5 The time from when I place my order until I receive my order is good enough for me
SEQ1 When I have a problem, the MFOA service shows a sincere interest in solving it
SEQ2 I feel safe in my transaction with the MFOA service in terms of security and privacy protection
SEQ3 The MFOA gives me individual attention
SEQ4 The MFOA service is always willing to help customers
SEQ5 The shippers of the MFOA take good care of the shipping job Self-developed
PPR1 The MFOA frequently offers mobile coupons
PPR2 The promotion activities offered by the MFOA meet my needs
PPR3 Mobile coupons and preferential schemes offered by the
MFOA are valuable to me
Various kind of discount benefits are offered when the MFOA is used (examples: mileage deposit, mobile coupons, ) Ye-Eun Song et al
PP1 The MFOA I am using is reasonably priced
PP2 The MFOA I am using is good value for the money
PP3 At the current price, the MFOA I am using provides good value
PP4 When I order food through the MFOA, the food is economical
VFC1 The MFOA offers a variety of food choices Yang et al (2004);
VFC2 The MFOA offers a variety of restaurant choices
VFC3 I can order food with a wide range of prices through the
VFC4 The MFOA offers food choices that meet my needs
(2019) VFC5 The MFOA frequently offers new restaurant choices
CS1 The MFOA has met my expectations Kohli et al (2004);
CS2 I strongly recommend this MFOA to others Lee & Chung (2009);
CS3 Overall, I am satisfied with the products/services of this
CS4 My choice to order foods from this MFOA was a wise one
CS5 I am happy with this MFOA
Continued Intention to Use (CI)
CI1 Assuming I want to order meals, I intend to reuse this
Ye-Eun Song et al (2017);
CI2 I will continue using this MFOA in the future
CI3 I intend to use this MFOA as frequently as now or more frequently in the future
CI4 I will preferentially consider the use of this MFOA over other MFOAs
Analysis Method
Data gathered from the online survey will be analyzed using Partial Least Squares Structural Equation Modeling (PLS-SEM) The research model and its associated hypotheses will be tested concurrently through structural equation modeling Prior to conducting confirmatory factor analysis to validate the proposed measurement indices, we will assess internal consistency reliability, convergent validity, and discriminant validity.
ANALYSIS RESULTS
Demographic Statistic
Table 4.1: Demographic Characteristic of Respondents
From VND 5 millions to 10 millions 48 17.0
From VND 11 millions to 20 millions 47 16.7
8 The MFOA that you have used the most
The survey revealed that the majority of participants were female, comprising 77.3% (218), while male participants accounted for 20.9% (59), and 1.8% (5) identified as other genders Most participants, 50.7% (143), were aged between 18 to 22 years, with 28.7% (109) in the 23 to 30 age range, and 10.6% (30) between 31 to 40 years; there were no participants over 40 A significant portion, 50.7% (143), resided in Hanoi, followed by 39.7% in Ho Chi Minh City, with smaller percentages living in Da Nang (4.6%) and other cities (5.7%) Regarding occupation, 63.5% (179) were students, 26.6% (75) were office staff, and 5% (14) were entrepreneurs or in other professions Income distribution showed that 56.4% (159) earned below 5 million VND, while 17% (48) earned between 5 to 10 million VND, 16.7% (47) earned 11 to 20 million VND, and 9.9% (28) earned over 20 million VND In terms of mobile food ordering app usage, 12.8% (36) had used them for less than six months, 16% (45) for six months to one year, 27.7% (78) for one to two years, 22.7% (64) for two to three years, and 20.9% (59) for over three years Nearly half, 48.9% (138), used the apps 1-3 times a month, while 15.2% (43) used them weekly, and 23.8% (67) used them 2-3 times a week The most popular apps were Now (40.4%), Baemin (27.3%), GrabFood (24.5%), GoFood (6.4%), and others (1.4%) Further details on demographics and purchasing behavior are available in the accompanying table.
Descriptive analysis
Table 4.2: Descriptive statistic of the scale items
Table above provides an overview of the descriptive statistics for al measurement items
The current study reveals a positive perception of MFOAs among users, highlighted by an average mean of 3.67 for information quality, with a standard deviation of 0.805 Participants also found MFOAs to be fairly priced, reflected in a mean value of 3.13 (1.01) for price perception Furthermore, perceived promotions garnered a strong positive impression, averaging 3.91 (0.92), while service quality was rated positively with an average mean of 3.66 (0.83) Users expressed favorable views on system quality and food variety, with averages of 3.94 (0.88) and 4.17 (0.85), respectively Additionally, respondents demonstrated a willingness to continue using MFOAs in the future, evidenced by a mean score of 3.84 (0.72) for continued intention items.
Common method bias test
Table 4.3: Common method bias Total Variance Explained
Harman’s single-factor test is a widely utilized method for assessing common method bias in self-reported surveys (Kock, 2020; MacKenzie & Podsakoff, 2012) The results of the test revealed that the first factor explained 25.820% of the total variance, which is below the 50% threshold suggested by MacKenzie and Podsakoff, indicating a low likelihood of common source bias.
(2012) Hence, there is no concern about the common method bias of the current study data.
Validity and reliability measurements
To evaluate the reliability and validity of the constructs, various measures were employed, including composite reliability (CR), Cronbach's alpha, average variance extracted (AVE), and discriminant validity, as outlined by Fornell and Larcker (1981) and Hair et al (2014) Additionally, convergent validity was assessed by examining the factor loading values for each scale item (Hair et al., 2014).
Reliability analysis
Cronbach's alpha is a widely used method for assessing the reliability of variables, with values above 0.70 indicating reliable constructs, as per Nunnally & Bernstein (1978) In this study, the reliability analysis revealed that the Corrected Item-Total Correlation for items "IQ5" and "CS2" fell below the acceptable threshold of 0.3, leading to their elimination After removing these items and re-evaluating, all remaining items achieved a Cronbach's alpha above 0.7 and Corrected Item-Total Correlation values exceeding 0.3, confirming the reliability of the scale The detailed results of the reliability analysis are presented in the table below.
Table 4.4: Summary Cronbach's alpha value of variables
Variable Code Number of items
Information Quality IQ 4 (rejected IQ5) 0.808
Customer Satisfaction CS 4 (rejected CS2) 0.856
Continued Intention to Use CI 4 0.854
Initially, Information Quality includes 5 measuring items and its original Cronbach’s alpha value is displayed in the table as follows:
Table 4.5: Initial Cronbach's alpha value of IQ – 1st test
Scale Mean if Item Deleted
Scale Variance if Item Deleted
Cronbach's Alpha if Item Deleted
The analysis indicates that item IQ5 has a Corrected Item-Total correlation of less than 0.3, necessitating its removal from the construct Following this adjustment, the Cronbach's alpha for the IQ construct was recalculated using the remaining four items (IQ1, IQ2, IQ3, IQ4), which met the reliability criteria for the construct.
Table 4.6: Cronbach's Alpha Value of IQ – 2nd test
Scale Mean if Item Deleted
Scale Variance if Item Deleted
Cronbach's Alpha if Item Deleted
Initially, Customer Satisfaction includes 5 measuring items and its original Cronbach’s alpha value is displayed in the table as follows
Table 4.7: Initial Cronbach's alpha value of CS - 1st test
Scale Mean if Item Deleted
Scale Variance if Item Deleted
Cronbach's Alpha if Item Deleted
The analysis indicates that item CS2 has a Corrected Item-Total correlation below 0.3, necessitating its removal from the construct Following this adjustment, Cronbach's alpha was recalculated for the remaining items (CS1, CS3, CS4, CS5), confirming that the construct for customer satisfaction (CS) now meets the reliability criteria.
Table 4.8: Cronbach's Alpha Value of CS - 2nd test
Scale Mean if Item Deleted
Scale Variance if Item Deleted
Cronbach's Alpha if Item Deleted
In this section, exploratory factor analysis (EFA) is utilized to evaluate the model's validity The KMO index is 0.877, exceeding the 0.5 threshold, while Bartlett's Test shows a significance level of 0.000, which is below the 0.05 requirement Additionally, the cumulative percentage of squared loadings stands at 58.297%, surpassing the 50% benchmark (Hair et al., 2014) These results affirm the data's suitability for exploratory factor analysis.
Table 4.9: Initial result of KMO and Bartlett's Test – 1st test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy 0.877 Bartlett's Test of Sphericity Approx Chi-Square 5179.182 df 595
Table 4.10: Initial result of Total Variance Explained – 1st test
Rotation Sums of Squared Loadings a
The Pattern matrix indicates that the factor loadings for SEQ3, VFC5, and SYQ3 are below 0.5, necessitating their removal from the measurement model and a subsequent re-evaluation of the Exploratory Factor Analysis (EFA).
Table 4.11: Initial result of Pattern Matrix – 1st test
After excluding the measurement items SEQ3, VFC5, and SYQ3 from the model, the results indicate that the KMO value is 0.870, exceeding the threshold of 0.5, while the significance level is 0.000, which is below 0.05 Additionally, the Total Variance Explained stands at 56.846%, surpassing the 50% benchmark.
Table 4.12: Result of KMO and Bartlett's Test - 2nd test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy 0.870 Bartlett's Test of Sphericity Approx Chi-Square 4097.410 df 496
Table 4.13: Result of Total Variance Explained – 2nd test
Rotation Sums of Squared Loadings a Total
The Pattern Matrix indicates that all factor loadings exceed 0.5, and after removing three items (SEQ3, VFC5, SYQ3), a total of 32 variables were organized into 8 distinct components.
Table 4.14: Result of Pattern Matrix – 2nd test
In conclusion, following the reliability analysis using Cronbach’s alpha and the Exploratory Factor Analysis (EFA), the model has been refined to include eight components: IQ, SYQ, SEQ, PPR, PV, VFC, CS, and CI Notably, the items IQ5, CS2, SEQ3, VFC5, and SYQ3 were excluded from the final model.
Testing Research model using PLS - SEM analysis
The study employed a two-stage approach using PLS-SEM to test both the measurement and structural models Initially, the measurement model was evaluated to ensure its quality, followed by an assessment of the structural model PLS, or Partial Least Squares, is a structural equation modeling technique that estimates relationships among constructs through correlation and a principal component method (Hair et al., 2014).
To ensure the reliability and validity of constructs, it is essential to evaluate several key metrics, including composite reliability (CR), Cronbach’s alpha, average variance extracted (AVE), and discriminant validity, as outlined by Fornell & Larcker (1981) and Hair et al (2014).
According to Hair et al (2014), for observed variables to be considered valid, their outer loading index must exceed 0.708, demonstrating that the construct accounts for over half of the variation in the indicator The results presented in the table below confirm that the measurement model meets these criteria, as all items exhibit outer loadings greater than 0.708, signifying strong relationships between the items and their respective constructs.
Table 4.15: Result of Outer loadings
CI CS IQ PPR PV SEQ SYQ VFC
The data presented in the table indicates that all constructs achieved Cronbach’s alpha and Composite Reliability values exceeding 0.7, in line with the recommendations of Fornell & Larcker (1981) and Hair et al (2014) Additionally, the Average Variance Extracted (AVE) values for all constructs were above the 0.50 threshold, further satisfying the criteria set by the same authors Consequently, it can be concluded that the constructs demonstrate convergent validity.
Table 4.16: Cronbach's Alpha, CR, AVE
Fornell and Larcker (1981) assert that discriminability is ensured when the square root of the Average Variance Extracted (AVE) for each latent variable exceeds the inter-correlation values Furthermore, it is essential that all square roots of AVE are greater than 0.5 The results indicate that all constructs meet these criteria.
CI CS IQ PPR PV SEQ SYQ VFC
Additionally, the HTMT values of the elements are all less than 0.85 (Henseler et al.,
2015) As a result, it can be concluded that all constructs demonstrated discriminant validity
Table 4.18: Heterotrait-Monotrait Ratio (HTMT)
CI CS IQ PPR PV SEQ SYQ VFC
Before testing the formulated hypotheses, the issue of multicollinearity was assessed According to Hair et al (2014), a Variance Inflation Factor (VIF) value of 3 or lower is optimal to prevent multicollinearity The analysis revealed that all VIF values in this study were below 3, with the highest value recorded at 1.265, indicating that multicollinearity is not a concern in the model The examination of VIF confirmed the absence of multicollinearity between the independent variables (IQ, SYQ, SEQ, PPR, PV, VFC) and the dependent variables (CS, CI).
CI CS IQ PPR PV SEQ SYQ VFC
The adjusted R square value for the dependent variable CI is 0.299, indicating that the independent variables account for 29.9% of the variation in CI In contrast, the adjusted R square value for the dependent variable CS is 0.440, meaning that the independent variables explain 44.0% of the variation in CS.
The study employed structural equation modeling (SEM) for hypothesis testing, utilizing a one-tailed approach for the directional research hypothesis The accompanying table presents the path coefficients along with their respective significance levels.
Table 4.22: Result of hypothesis testing
P Values f square Decision H1 IQ -> CS 0.058 1.105 0.135 0.005 Not Supported H2 SYQ -> CS 0.290 4.848 0.000 0.127 Supported
H5 PV -> CS -0.022 0.486 0.314 0.001 Not Supported H6 VFC -> CS 0.345 5.666 0.000 0.197 Supported
All hypotheses are one-tail tests;
The path coefficient analyses indicated that Customer Satisfaction is significantly influenced by several factors: SYQ (β = 0.290, p < 0.05), Service Quality (β = 0.157, p < 0.05), Perceived Promotions (β = 0.250, p < 0.05), and VFC (β = 0.345, p < 0.05) In contrast, Information Quality (β = 0.058, p > 0.05) and Price Value (β = -0.022, p > 0.05) did not have a significant impact on Customer Satisfaction Furthermore, Customer Satisfaction was found to have a positive effect on Continued Intention to Use (β = 0.549, p < 0.05).
Additional Analysis – Mediating Effect
There are two hypotheses are rejected; therefore, the author proceeds to investigate if any mediator exists in the model
Figure 4.3: PLS - SEM diagram (2) Table 4.23: Result of Mediating Effect
Indirect-only mediation (Full mediation)
The study reveals that Information Quality does not significantly influence Continued Intention to Use, as evidenced by a direct effect coefficient of β= -0.045 (p>0.05) and an indirect effect through Customer Satisfaction with a coefficient of β= 0.017 (p>0.05) Consequently, there is no direct or mediating effect of Information Quality on Continued Intention to Use.
The data presented in the table indicates that System Quality does not have a direct effect on Continued Intention to Use, with a coefficient of β= 0.130 and a p-value greater than 0.05 However, it does exert an indirect influence on Continued Intention to Use through Customer Satisfaction, demonstrated by a coefficient of β= 0.084 and a p-value less than 0.05, as noted by Hair et al.
In 2014, it was established that when the indirect effect is significant while the direct effect is not, this indicates full mediation, meaning the indirect-only effect is present Conversely, when the direct effect is significant and the indirect effect is not, it suggests partial mediation, or a direct-only effect In the context of the relationship between System Quality and Continued Intention to Use, Customer Satisfaction serves as a full mediator.
Service Quality significantly influences Continued Intention to Use, both directly (β= 0.179, p