NGUYỄN THỊ THUẬN AN-1701015008 -KTĐN

FOREIGN TRADE UNIVERSITY
HO CHI MINH CITY CAMPUS

---------***--------
GRADUATION THESIS
Major: International Business Economics
FACTORS AFFECTING THE CONTINUANCE

INTENTION TO USE FOOD DELIVERY APPS OF
THE MILLENNIALS IN HO CHI MINH CITY
Student: Nguyễn Thị Thuận An

Student ID: 1701015008
Class: K56CLC3
Intake: 56
Supervisor: Mr. Lê Giang Nam
Thesis ID: 302
Ho Chi Minh City, December 18th, 2020

FOREIGN TRADE UNIVERSITY SOCIALIST REPUBLIC OF VIETNAM
HO CHI MINH CITY CAMPUS Independence - Freedom - Happiness
Ho Chi Minh City, ….../….../...............
DISSERTATION REMARKS
Student’s full name: NGUYỄN THỊ THUẬN AN Student code:

1701015008
Subject of the thesis: Factors affecting the continuance intention to use food
delivery apps of the Millennials in Ho Chi Minh city.
Name of the supervisor: Mr. Lê Giang Nam
Department: Foreign Trade University Ho Chi Minh City Campus
DETAILED REMARKS
1. Student’s attitude and progress during the dissertation schedule:
.......................................................................................................................................
.......................................................................................................................................
2. Responsiveness of scientific content to the research topic:
.......................................................................................................................................
.......................................................................................................................................
.......................................................................................................................................
3. Format of the dissertation:
.......................................................................................................................................
.......................................................................................................................................
4. The supervisor proposes the score for student’s attitude and progress during
the dissertation schedule:
.........../1
Lê Giang Nam
STATEMENT OF AUTHORSHIP
Except where reference is made in the text of the thesis, this thesis contains no
material published elsewhere or extracted in whole or in part from a thesis by which
I have qualified for or been awarded another degree or diploma.
No other person’s work has been used without due acknowledgements in the
thesis.
This thesis has not been submitted for the award of any degree or diploma in
any other tertiary institution.
Ho Chi Minh City, December 2020
Nguyễn Thị Thuận An

ACKNOWLEDGMENT
First and foremost, the author would like to express her deepest gratitude to
her supervisor – Mr. Le Giang Nam. Without his kind direction and wholehearted
guidance, this graduation thesis would have been a little success. In every phase of
the project, his supervision and guidance shaped this thesis to be as flawless as
possible.
Secondly, the author wishes to render her special thanks to four experts,
together all individuals engaged in focus group interviews, for their thorough review
as well as constructive feedbacks so as to revise her measurement scales and survey
questionnaire. Besides, the author is ever grateful to all respondents, who spared their
precious time completing the survey so that the research can be accomplished.
Despite great effort, due to the author’s restricted knowledge and self-
capabilities, this thesis inevitably contains some shortcomings. Thus, the author
hopes to receive constructive feedbacks from lecturers and professionals to perfect
the thesis’s academic content and improve for future research. Finally, the author also
would like to take this opportunity to wish all lecturers from Foreign Trade University
Ho Chi Minh Campus health, happiness and success in work and life.
Ho Chi Minh City, December 2020
Nguyễn Thị Thuận An

TABLE OF CONTENT
LIST OF ABBREVIATIONS
LIST OF TABLES AND FIGURES
CHAPTER 1: INTRODUCTION ............................................................................ 1
1.1. Rationale of the research ............................................................................... 1
1.2. Research aims and objectives........................................................................ 3
1.2.1. Research aims ...........................................................................................3
1.2.2. Research objectives ...................................................................................4
1.3. Subject and scope of the research ................................................................. 4
1.3.1. Research subject ........................................................................................4
1.3.2. Research scope ..........................................................................................5
1.4. Research background .................................................................................... 5
1.4.1. Foreign studies ..........................................................................................5
1.4.2. Domestic studies .......................................................................................6
1.5. Research questions ............................................................................................. 6
1.6. Research methodology ....................................................................................... 7
1.7. Contribution and significance of the research ................................................ 8
1.8. Research structure ............................................................................................. 9
SUMMARY OF CHAPTER 1 ............................................................................. 9
CHAPTER 2: THEORETICAL FRAMEWORK AND LITERATURE

REVIEW .................................................................................................................. 10
2.1. Food delivery ................................................................................................ 10
2.1.1. The concept of food delivery ..................................................................10
2.1.2. Food delivery market in Vietnam ...........................................................11
2.2. Food delivery app (FDA) ............................................................................. 13
2.2.1. Food delivery app....................................................................................13

2.2.2. Classification of food delivery app ........................................................ 14
2.2.3. Food delivery app market in Vietnam .................................................... 14
2.2.4. Food delivery app usage in Vietnam ...................................................... 17
2.3. The Millennials .............................................................................................18
2.3.1. Definitions of the Millennials ................................................................ 18
2.3.2. Some key characteristics of the Millennials .......................................... 19
2.4. The concept of behavioral continuance intention to use ..........................21
2.4.1. Behavioral intention to use..................................................................... 21
2.4.2. Behavioral continuance intention to use (CI) ........................................ 22
2.5. Conceptual models and theories on factors affecting continuance

intention to use food delivery apps ....................................................................24
2.5.1. Theory of Reasoned Action (TRA) (1975) ............................................ 24
2.5.2. Technology Acceptance Model (TAM) (1989) ..................................... 24
2.5.3. Task-Technology Fit (TTF) Theory (1995) ........................................... 25
2.5.4. Unified Theory of Acceptance and Use of Technology (UTAUT) (2003)

.......................................................................................................................... 26
2.5.5. Unified Theory of Acceptance and Use of Technology 2 (UTAUT2)

(2012) ............................................................................................................... 26
2.5.6. Expectancy Confirmation Model (ECM) (2001) ................................... 27
2.6. Empirical research on continuance intention to use food delivery apps 28
2.6.1. Foreign studies ....................................................................................... 28
2.6.2. Domestic studies .................................................................................... 30
2.7. Hypothesis development and proposed research model ...........................30
2.7.1. Rationale for the proposed research model ............................................ 30
2.7.2. The proposed research model................................................................. 32
2.7.3. Hypothesis development ........................................................................ 32

SUMMARY OF CHAPTER 2 ........................................................................... 41
CHAPTER 3: RESEARCH METHODOLOGY ................................................. 42
3.1. Research process .......................................................................................... 42
3.2. Preliminary research ................................................................................... 43
3.2.1. Initial questionnaire development ...........................................................43
3.2.2. Pretest: Expert reviews and focus groups ...............................................45
3.2.3. Pilot study................................................................................................47
3.2.4. Official questionnaire ..............................................................................47
3.3. Official research ........................................................................................... 47
3.3.1. Data collection and sample size determination .......................................47
3.3.2. Data analysis ...........................................................................................48
CHAPTER 4: RESEARCH RESULTS ................................................................ 53
4.1. Descriptive statistics analysis ...................................................................... 53
4.1.1. General description .................................................................................53
4.1.2. Statistical description ..............................................................................53
4.2. Cronbach’s Alpha coefficient analysis ....................................................... 55
4.3. Explanatory Factor Analysis (EFA) ........................................................... 56
4.4. Confirmatory Factor Analysis (CFA) ........................................................ 57
4.5. Structural Equation Modeling (SEM)........................................................ 59
4.6. Independent Sample t-Test & One-Way ANOVA on demographic

variables ............................................................................................................... 62
4.7. Research result discussion ........................................................................... 63
4.7.1. Research result ........................................................................................63
4.7.2. Research result discussion ......................................................................64

CHAPTER 5: RESEARCH CONCLUSION AND RECOMMENDATIONS ..69
5.1 Research conclusion ......................................................................................69
5.2. Research recommendations.........................................................................69
5.2.1. Recommendations for food delivery app providers ............................... 69
5.2.2. Recommendations for food & beverage partners on food delivery

platforms........................................................................................................... 77
5.2.3. Recommendations for traditional restaurants ........................................ 77
5.2.4. Recommendations for policy makers ..................................................... 78
5.3. Limitations and proposed future research orientation ............................79
SUMMARY OF CHAPTER 5 ...........................................................................80
REFERENCES
APPENDICES
LIST OF ABBREVIATIONS
No. Abbreviation Meaning

1 AMOS Analysis of Moment Structures
2 ANOVA Analysis of Variance
3 ASEAN Association of Southeast Asian Nations
4 AVE Average Variance Extracted
5 CFA Confirmatory Factor Analysis
6 CFI Comparative Fit Index
7 CI Continuance intention
8 CMIN Chi-square (x2)
9 CMIN/df Chi-squared adjusted for degrees of freedom
10 COF Confirmation
11 CR Composite Reliability
12 ECM Expectancy Confirmation Model
13 ECT Expectation-Confirmation Theory
14 EE Effort expectancy
15 EFA Exploratory Factor Analysis
16 Estimates Standardized Loading Estimates (Factor Loading)
17 Etc. et cetera
18 F&B Food and Beverage
19 FC Facilitating condition
20 FDA(s) Food delivery app(s)
21 GFI Goodness of fit index
22 HB Habit
23 HCM city Ho Chi Minh City
24 HM Hedonic motivation
25 IS Information system
26 i.e. id est (meaning: That is)
27 KMO Kaiser – Meyer – Olkin
28 MFOA Mobile food ordering apps
29 MSV Maximum Shared Variance
30 PCD Platform-to-Consumer Delivery
31 PCLOSE P-value for a test of close fit
32 PE Performance expectancy
33 PhD Doctor of Philosophy
34 PO Price-saving Orientation
35 R2 Squared coefficient of multiple correlation
36 RCD Restaurant-to-Consumer Delivery
37 RMSEA Root Mean Square Errors of Approximation
38 SA Satisfaction
39 SEM Structural Equation Model
40 SI Social influence
41 Sig. Significant
42 SPSS Statistical Product and Services Solutions
43 SQRTAVE Square Root of AVE
44 TAM Technology Acceptance Model
45 TLI Tucker-Lewis Index
46 TRA Theory of Reasoned Action
47 TTF Task-Technology Fit
48 USD United States Dollar
49 UTAULT Unified Theory of Acceptance and Use of Technology
Unified Theory of Acceptance and Use of Technology
50 UTAUT2
2
51 VIF Variance Inflation Factor
52 VND Vietnamese Dong
LIST OF TABLES AND FIGURES
List of tables
No. Table Name of table Page
Revenue of Vietnamese food delivery market during

1 Table 2.1 12
2017 – 2024 period
2 Table 3.1 Measurement scales for all variables 43
3 Table 3.2 Adequacy of Cronbach's Alpha value 49
4 Table 3.3 Criteria for assessment CFA/SEM statistical results 51
EFA analyzing procedure and evaluation criteria for

5 Table 4.1 56
CFA/SEM-based model
6 Table 4.2 KMO, Bartlett's Test and Total Variance Explained 57
7 Table 4.3 Model-fit indices of measurement model 58
8 Table 4.4 Standardized Regression Weights 61
9 Table 4.5 Squared Multiple Correlations 62
Economic, social and environmental impacts of food

10 Table 5.1 78
delivery sector
List of figures
No. Figure Name of figure Page
Process of ordering and delivery food via food

1 Figure 2.1 13
delivery apps
2 Figure 2.2 Task-Technology Fit (TTF) Theory (1995) 25
Unified Theory of Acceptance and Use of

3 Figure 2.3 27
Technology 2 (UTAUT2) (2012)
4 Figure 2.4 Expectancy Confirmation Model (ECM) (2001) 28
5 Figure 2.5 The proposed research model 32
6 Figure 3.1 Research process 42
7 Figure 3.2 Procedures for testing one-way ANOVA 52
8 Figure 4.1 Result of research model 64

1
CHAPTER 1: INTRODUCTION
1.1. Rationale of the research
Within the fourth industrial revolution, technology has been ingrained into
almost every single facet of human lives, and the food and beverage industry is no
exception (Deloitte, 2019). Online delivery services are reshaping the industry
globally and locally since the number of consumers ordering food online in Vietnam
skyrockets (Google and Temasek, 2019; Facebook and Bain & Company, 2020).
In fact, Vietnam’s food delivery market has experienced spectacular growth

over the last few years. According to the market research company Euromonitor
International, food delivery market in Vietnam was valued at USD 33 million in 2018.
This figure is expected to reach USD 38 million in 2020 and will sustain an average
growth of 11% in the next 5-year period. Besides, based on the market research
company Statista’s reports, Vietnam’s revenue in the food delivery market is
projected to amount up to USD 302 million in 2020. Revenue is expected to witness
a compound annual growth rate of 16.5% during the 2020-2024 period, resulting in a
market volume of USD 557 million by 2024. Despite this huge market value,
compared to other Southeast Asian countries such as Indonesia and Singapore, the
size of this market in Vietnam is still very small (Google and Temasek, 2019). Yet,
thanks to that, Vietnamese food delivery market is considered a "golden" market that
attracts a plethora of domestic and foreign investors.
Among the predominant food ordering and delivery methods, food delivery
applications (FDAs) emerge as the most popular one, especially in the large
metropolises such as Ho Chi Minh City (HCM city) and Hanoi (Q&Me, 2020). This
finding is in line with the major rate of smartphone ownership (Q&Me, 2020),
together with the growing trends of mobile application usage and e-commerce
spending in food category within Vietnamese market (We Are Social and Hootsuite,
2020). Some of the leading FDA providers operating within this segment include
GrabFood, Now, GoFood, Baemin, Loship (Q&Me, 2020). With the exception of
Loship as a domestic startup, the other platforms are all backed by foreign investors.
2
Additionally, food delivery market is not only the playground of FDA

providers, but also a profitable landscape for food and beverage (F&B) merchants,
who register their food stall to provide food and/or drink via these platforms.
Apparently, food delivery platforms enable merchant partners to reduce their expense
and streamline several cumbersome processes, especially renting location, spending
on advertising costs, and developing shipper workforce. Additionally, F&B
merchants can receive consultancy from food delivery companies - in terms of market
research, packaging, delivery, to name a few – so as to optimize their selling process.
As a pioneer in food delivery market, Now has attracted over 20,000 stores
providing food and drink together with countless attractive promotions on the
platform (Vietnam Credit, 2020). Concerning GrabFood, it has expanded its network
of more than 8,500 F&B merchants during the short span of two years after its launch
in 2018, diversifying its options ranging from fast food to bubble tea and mala hotpot
(Vulcan Post, 2020). Statistics of D’ corp R-Keeper Vietnam show that Vietnam’s
F&B industry has around 540,000 eateries, including 22,000 coffee lounges, 7,000
fast food outlets, and more than 80,000 facilities run by large F&B chains. If merely
3% of the nation’s restaurants go online, that would add another 19,000 merchants to
delivery platforms, presenting a huge boom in the delivery market.
In the light of COVID-19 pandemic and governmental orders for social

distancing, food delivery service in Vietnam has been on a much more meteoric rise
(Q&Me, 2020). Survey by ride-hailing firm GoJek reveals that food delivery demand
has risen sharply. From February 2nd to 9th 2020, there were over 650,000 food orders
on the platform, with the numbers increasing every day.
As regards consumers, for a large number of people, ordering food online has
become a natural habit. There has been a remarkable change in consumers’ lifestyle
owning to rising urbanization and busy work schedules, which has obligated them to
opt for convenient food delivery options (IMARC, 2019). This is especially true when
it comes to the Millennials (those born from 1980 to 2000), who are time-starved and
convenience-seeking consumers (My Nguyen, 2019; Intage Vietnam, 2020).
Specifically, recent surveys highlight that a large percentage of respondents order
3
foods through FDA apps at least once a week (GCOMM, 2020; Q&Me, 2020). Such
a trend tends to escalate during and even after COVID-19, as the eating out frequency
decreased in account of health concerns (Nielsen, 2020; Zhao and Bacao, 2020).
In view of the growing usage patterns as well as thriving delivery market with
the presence of a handful of FDA providers (Research and Market, 2018; Facebook
and Bain & Company, 2020), it is a prerequisite to figure the exact drivers behind
consumers’ behavioral intention to continuously use these food delivery apps.
However, given that online food delivery apps have been attracting significant
interest in Vietnam and other countries in the Southeast Asian region, the related
issues of these apps have not been fully studied by academics and researchers.
Additionally, most previous studies of mobile apps in general and FDAs in particular
have simply addressed aspects related to intention to use and initial adoption of
Vietnamese people. Although first-time use is a crucial indicator of information
system success, it does not necessarily promote the desired managerial outcome
unless such use does continue in the longer term (Lyytinen and Hirschheim, 1987;
Bhattacherjee, 2001; Limayem et al., 2003; Kim and Malhotra, 2005). Therefore, this
study entitled “Factors affecting the continuance intention to use food delivery
apps of the Millennials in Ho Chi Minh city” will go further by examining
customers’ satisfaction and their continuance intention to use, especially when most
FDAs have been already popular and well adopted by customers. From that, some
viable managerial implications and recommendations can be proposed so as to ignite
future growth of FDAs and food delivery market as a whole.
1.2. Research aims and objectives

1.2.1. Research aims
Acknowledging the strong growth potential of food delivery services, with the
escalating usage pattern of FDAs, especially within the tech-savvy and convenience-
driven Millennials, the author conducted this research so as to portray customer
insight and figure out factors driving their FDA usage continuance intention.
To be specific, this research aims to (1) determine factors affecting the

continuance intention to use FDAs of the Millennials in HCM city, (2) examine the
4
level of influence that model constructs have on customers’ continuance intention,

and from which (3) propose recommendations tailored for key market participants
(FDA providers, F&B merchants, traditional restaurateurs, policy makers).
1.2.2. Research objectives

This research is conducted with three principal objectives as follows:
Firstly, the research systematized theoretical basis of customers’ behavioral

intention in general, and their continuance intention to use FDAs in particular. In
addition, the author presented an overview of food delivery market and FDA usage
in Vietnam, including some impacts of the past COVID-19 epidemic. The author also
synthesized and evaluated the findings of previous studies, from which the research
model and hypotheses were proposed accordingly.
Secondly, based on the proposed research model, an in-depth questionnaire

was set up to measure the influence level of each given model construct along with
their relationships. Once pre-test steps, pilot study and online convey had been
conducted, the author performed data analysis and testing, leveraging the powerful
structural equation modeling software IBM SPSS Statistics 20.0 and IBM SPSS
AMOS 24.0.
Thirdly, from the conclusions drawn from the statistical findings, the author
proposed some key solutions and recommendations to enhance customers’
satisfaction and foster their intention to continuously use FDAs.
1.3. Subject and scope of the research

1.3.1. Research subject
Research subject: Factors affecting the continuance intention to use FDAs of
the Millennials in HCM city.
Research respondents: The Millennials currently living in HCM city, those

born from 1980 to 2000, who have used FDAs. With their hectic schedules and
lifestyles, the Millennial population is considered a key driver of the profitability of
FDAs, being the more regular users of these apps compared to the Generation X and
Baby boomers.
5
1.3.2. Research scope

Scope of content: Generally, FDAs could be categorized into two types (Yeo
et al., 2017; Ray et al., 2019). The first one is branded delivery apps, which are
developed by food retailers themselves and popular among fast food chains. The latter
refers to third-party delivery platforms, or typically known as FDA providers,
including GrabFood, Now, GoFood or Baemin. The research focuses on the second
category, given its high and ever-increasing adoption rate of urban consumers
(Q&Me, 2020).
Scope of geography: Ho Chi Minh city. HCM city is among the key markets
for food delivery services, with total number of daily online food orders six times
higher than that in Hanoi as of 2019, according to surveys by GoJek (2019) and
Kantar (2019).
Scope of time:
(1) Primary data: Due to timing limitation of this graduation thesis, primary data in
this research were collected during the period from October 2020 to November 2020.
(2) Secondary data (numerical data): ranging from 2017 to 2020.
1.4. Research background

Despite the fact that FDAs have recently attracted much attention and widely
adopted globally and nationally, academic interest in examining topics related to
these platforms is still in its infancy (Wang et al., 2019; Ray et al., 2019; Alalwan,
2020).
1.4.1. Foreign studies

From a global perspective, the number of official papers focusing specifically
on FDAs is still limited, although there exist many empirical studies conducted both
on mobile apps and food delivery services in general, which, though, deliver great
reference values for the author’s research. Furthermore, whereas there are already
just a modest number of studies on FDA, the majority of them did focused solely on
intention to use, or the initial adoption of these platforms, although FDAs have been
long popular and widely adopted worldwide. Just recently, there are a few papers
6
examining consumers’ latter stages of FDA adoption, such as satisfaction,

continuance use intention or advocacy. These include the research by Kang and
Namkung (2018) (specifically for coffee branded apps), Lee et al. (2019), Alalwan et
al. (2020) and Zhao and Bacao (2020) (considering the context of COVID-19).
Notwithstanding that fact that these studies provide valuable theoretical
contributions, important gaps do exist. For instance, the research “Mobile food
ordering apps (MFOA): An empirical study of the factors affecting customer e-
satisfaction and continued intention to reuse” by Alalwan et al. (2020) employed a
comprehensive research model, extending Unified Theory of Acceptance and Use of
Technology 2 (UTAUT2) with three variables of online review, online rating, online
tracking. While the research findings were quite satisfactory (06 out of 10 variables
were validated), the given model did not consider some decisive technical and
cultural factors, typically mobile interactivity, food habits or lifestyle.
1.4.2. Domestic studies

Established in mid-2015 yet only starting to grow by 2018, food delivery is
still a new market in Vietnam (B & Company, 2019). Whilst food delivery market
and FDAs specifically have become prevalent over recent years, domestic studies
covering these topics have been rare. Especially when it comes to FDAs and FDA
users’ continuance intention to use, there is literally no previous research which has
been officially published so far. Given that, a few relevant papers examining related
topics, such as one by My Nguyen (2019) on food delivery services, can be utilized
as a source of reference.
All things considered, the author decided to perform an in-depth research to

study determinants driving continuance intention to use FDAs among the Millennials
in HCM City. Within this research, well-proven factors from previous studies,
combined with the once-neglected potential ones, are all taken into great account and
leveraged accordingly.
1.5. Research questions

Based on its aims and objectives, this study focuses on three principal
questions:
7
(1) What are the factors affecting the continuance intention to use food delivery apps
of the Millennials in Ho Chi Minh city?
(2) At which levels and from which side do these factors affecting the continuance
intention to use food delivery apps of the Millennials in Ho Chi Minh city?
(3) Which viable proposals can be put forth to some key market participants so as to
enhance the continuance intention to use food delivery apps of the Millennials in Ho
Chi Minh city?
1.6. Research methodology

In this thesis, the author compromised both qualitative and quantitative
methods by synthesizing primary and secondary data, which was demonstrated as
follows:
Qualitative research: This research was conducted through systemizing

empirical findings of previous studies, both foreign and domestic ones, along with
garnering data on related topics from reliable sources, including high-ranking science
journals and market research companies. On such basis, the author decided to
leverage an integrated model of UTAUT2, ECM and TTF while also developing 18
hypotheses. Two pretest techniques of expert review and focus group interview were
also adopted, aiming to acquire more insightful data and feedback for the following
quantitative research.
Quantitative research: Quantitative research is considered the most

commonly used method to analyze relationships among variables and test statistical
hypotheses in an experimental study (Cooper and Schindler, 2014; Bryman and Bell,
2015). This research was performed through conducting an online survey on the
questionnaire, which was built and adjusted in the preliminary research stage. Once
data had been collected, the author first tested reliability of measurement scales by
Cronbach's Alpha coefficient, then ran Explanatory Factor Analysis (EFA) to all
constructs. After that, Confirmatory Factor Analysis (CFA) and Structural Equation
Modeling (SEM) was carried out. In final step, t- test and Analysis of Variance
8
(ANOVA) were employed to assess the influence of demographic variables on

continuance intention of FDA users.
1.7. Contribution and significance of the research

1.7.1. Contribution of the research
Firstly, in Vietnam, whereas there are many studies covering the topics of
online shopping or the adoption of mobile applications, in-depth research on FDAs
and intention to use and/or reuse these apps is still limited. Even when there exist
some academic studies focusing specifically on FDAs, they typically addressed
aspects related consumers’ initial adoption. Hence, given that FDAs have already
well adopted in Vietnam, it is essential to perform a research on FDA users’
continuance usage intention.
Secondly, rather than being conducted on general customers as most previous

studies, this research focuses on the Millennials, a particular age group with
distinctive characteristics, shopping habits and motivations (European Union, 2020).
As this generation proves to be the drivers for food delivery services, their crystal-
clear insights when deciding to use FDAs continuously are undoubtedly valuable.
Thirdly, the author undertook this research by leveraging an integrated model

of UTAUT2, ECM and TTF, while also adjusting some variables of the original
model to make it more compatible with the research subject and target respondents.
So far, this is the first time this specific model has been employed, whether in
Vietnam or all over the world, whether for FDAs or mobile apps in general. Thus,
this research will constitute a new theoretical contribution, hopefully filling gaps in
existing empirical studies.
1.7.2. Significance of the research

Examining factors shaping FDA users’ continuance usage intention – a critical
indicator for long-term viability and profitability of FDAs (Bhattacherjee, 2001), this
research aims to present some practical implications for key stakeholders involved.
For FDA providers and F&B partners, they can obtain deeper insights into
Millennials’ consumers, their current experiences with FDAs as well as key
9
determinants driving their decision to continuously use FDAs. Understanding that,

these partices can adapt their strategies and align marketing campaigns accordingly
to earn users’ loyalty.
For traditional restaurants, whose sole business is in their physical storefront,

they can consider revamping business model, integrating delivery into their daily
operation and adopting an omni-channel approach.
For policy-makers, they can formulate some incentive programs to positively

support consumers (FDA users), FDA providers, F&B merchants and traditional
restaurants, aiming to accelerate the growth of food delivery sector. Towards an effort
to facilitate such growth, they should take into account all impacts of this sector and
establish proper course of action to address them well.
1.8. Research structure

This research is structured into 05 chapters as below:
Chapter 1: Introduction
Chapter 2: Theoretical Framework and Literature Review
Chapter 3: Research Methodology
Chapter 4: Research Results
Chapter 5: Research Conclusion and Recommendations
SUMMARY OF CHAPTER 1
In this chapter, the author clarifies the research rationale to investigate factors
affecting continuance intention to use food delivery apps among the Millennials in
Ho Chi Minh city. A comprehensive and profound understanding over factors
shaping FDA users’ continuance use intention is expected to allow FDA providers as
well as their F&B merchants identify key business chances to focus on and then ignite
their success. Besides, after reading through several foreign and domestic studies, the
author hopes this research may make some valuable contributions to empirical
research literature on FDAs, delivering workable theoretical and managerial
implications.
10
CHAPTER 2: THEORETICAL FRAMEWORK AND LITERATURE

REVIEW
2.1. Food delivery
2.1.1. The concept of food delivery
As there has been no official market definition of food delivery published by
governmental agencies so far, only a few temporary definitions are proposed by
business analysts and market research firms. Statista defines food delivery as services
in which meals, once ordered online, will be prepared and delivered for direct
consumption. Besides, based on the global market research IMARC Group, food
delivery can be understood as the process of ordering food from a food cooperative
or restaurant via webpage or mobile app. Likewise, Charlene Li at al. (2020) also
described food delivery as the process of preparing and/or delivering ordered food,
in which they attributed its prevalence to the development of integrated online food
delivery platforms. Drawn from the above-mentioned perspectives, in this research,
food delivery can be generally understood as a service or a process in which food
item is ordered over distance and then delivered to consumers’ doorsteps.
In terms of classification, food delivery is categorized into two segments,

consisting of Restaurant-to-Consumer Delivery (RCD) and Platform-to-Consumer
Delivery (PCD) (Statista, 2018). In RCD segment, food delivery is carried out directly
by restaurateurs, whether the order being made via platforms or directly through
restaurant website. The most well-known examples of this segment are such fast-food
chains as McDonalds, Burger King, Domino’s.
As regards the PCD, this market segment focuses on online delivery services
which provide customers with meals from partner restaurants that do not necessarily
have to offer food delivery themselves. According to the management consulting firm
McKinsey & Company, PCD services are a combination of order-focused food
delivery services (or food order aggregators) and logistics-focused food delivery
services (or “new delivery”). FDAs, as the main focus of this research, belong to this
segment - or more specifically, they are classified as new-delivery players.
11
2.1.2. Food delivery market in Vietnam

2.1.2.1. Food delivery market
Since the 20th century, food delivery has been established as a business
segment, with a handful of delivery services, especially for Chinese cuisine. As in
other sectors, huge waves of technology and digitalization have disrupted the market,
in which food delivery now can be applied to any kinds of food. The emergence of
PCD services offered by third-party companies have also driven the market to a far
more meteoric rise.
According to Statista, revenue generated within the food delivery landscape,

which is already adjusted for the expected impact of COVID-19, is anticipated to
reach USD 136,431 million in 2020. In the next 2020 – 2024 period, the revenue is
projected to experience an annual growth rate of 7.5%, resulting in an estimated
market volume of USD 182,327 million by 2024. Besides, compared to RCD, PCD
emerges as a larger segment, having a projected market volume of USD 70,741
million in 2020. Appendix 1 presents the market outlook of online delivery segment
in major countries where food delivery market has been well-established.
Despite adopting food delivery much later than other regions, Asia Pacific was
considered the largest region in the online food delivery services market in 2019
while North America remained the second (Research & Markets, 2020). Being a part
of Asia Pacific, Southeast Asia has witnessed rapid expansion of its food delivery
business since 2015. As anticipated by Statista (2020), the region's food delivery
market would reach notably USD 3,492 million in 2020 and have the projected
market volume of USD 5,682 million by 2024.
2.1.2.2. Food delivery market in Vietnam
Globally ranked 32nd in terms of revenue (Statista, 2020), Vietnam stands
among the top potential food delivery markets around the world, witnessing
significant growth over the recent years. Based on Euromonitor International’s report,
food delivery market in Vietnam was valued at USD 33 million in 2018. This figure
is expected to reach USD 38 million in 2020 and will sustain an average growth of
11% in the next 5-year period. According to Statista, Vietnam’s revenue generated in
12
food delivery market is projected to amount up to USD 302 million by 2020. Besides,
revenue is expected to welcome a compound annual growth rate of 16.5% during the
period of 2017 – 2024.
Table 2.1: Revenue of Vietnamese food delivery market during 2017 – 2024 period
Unit: million USD
Year 2017 2018 2019 2020 2021 2022 2023 2024
Online delivery 102.0 147.8 207.1 302.1 376.9 446.3 506.7 556.8
market
PCD 17.7 31.5 51.2 84.1 112.1 138.9 162.8 183.0
RCD 84.3 116.3 155.9 218.0 264.8 307.4 343.9 373.8
Source: Statista, 2020

Both two types of food delivery enjoy continuous increase in users’ adoption
as well as strong revenue growth. Specifically, the number of users in RCD and PCD
segments were 6.2 and 1.4 million in 2019, which are all expected to be over two
times greater by 2024 (12.8 and 4.6 respectively). Regarding to revenue, while
revenue generated in RCD is higher than that of PCD, the latter has greater revenue
growth, as measured and anticipated for the 2017-2024 period (see Appendix 2).
Such strong growth of food delivery within Vietnam market can be attributed
to dramatic changes in consumers’ lifestyles and living habits, which resulted from
high speed of urbanization (specifically, 36% by January 2020) and their hectic
schedule (IMARC, 2019). Increasingly time-poor consumers value convenience ever
than before, which creates greater demand for technology-based delivery services to
save time (Deloitte, 2019). Furthermore, Vietnam has experienced significant rise in
Internet penetration, 70% by January 2020 compared to 58% in January 2018, with
around 68.17 million Internet users. Another critical factor that drives this market is
the rising rate of smartphone adoption in Vietnam, in which almost 93% of Internet
users aged 16 to 64 own a smartphone (We Are Social and Hootsuite, 2020).
Regarding methods of ordering food for delivery, the recent survey conducted
by Q&Me among 840 respondents in HCM city and Hanoi suggests that there are
four most popular food ordering methods in Vietnam, including (1) third-party
13
delivery apps, (2) dedicated apps of the store, (3) telephone and (4) social networking
sites. Among them, third-party delivery apps (FDAs) are proved the most popular,
with 94% respondents adopting them. As for the popularity of these methods by
region, FDA is more prevalent in HCM city while Hanoi has the higher ratio of social
media or telephone orders.
2.2. Food delivery app (FDA)

2.2.1. Food delivery app
FDA can be referred to as a mobile application that smartphone users adopt as
a convenient and innovative channel to access food stalls, view food menus, place
meal order(s), and make payments without any physical interaction with onsite
restaurant staff (Okumus and Bilgihan, 2014). Booming food delivery demand has
also resulted in expansion opportunity for ride-hailing apps as these services are
typically integrated into one single platform (Google and Temasek, 2019).
Below is a typical process of ordering and delivery food via FDAs.
Figure 2.1: Process of ordering and delivery food via food delivery apps
Source: Derived by the author, based on Li et al., 2020
One obvious advantage of utilizing FDAs is value addition of comfort and

convenience which makes many customers opt for it. Particularly, with FDA(s)
14
installed in their smartphones, users are able to access and order their food from a
wide range of restaurants at times and locations convenient to them. Additionally,
these FDAs offer customers with more comprehensive, up-to-date, and accurate
information about various restaurants as well as their menu options. Accompanying
this information is the ability for FDA users to track their order progress throughout
its stages (Lu and Rastrick, 2014; Nilashin et al., 2015; Algharabat et al., 2017).
Besides saving coupons, FDAs are embraced for a myriad of tech-based innovative
features that allow both users and restaurants to address annoying problems, typically
long waiting times in stores.
2.2.2. Classification of food delivery app

Quite similar to food delivery services, FDAs could be categorized into two
types (Yeo et al., 2017; Ray et al., 2019). The first is branded delivery apps, which
are developed by restaurants or food retailers themselves. In Vietnam, this type is
popular among fast food chains, such as Pizza Hut, McDonalds, Domino's Pizza, as
well as coffee shops, such as Highlands Coffee, Starbucks, The Coffee House. The
latter refers to third-party delivery platforms, or better known as FDA providers.
These platforms vary from country to country, and in Vietnam, they are comprised
of GrabFood, Now, GoFood, Baemin, Loship. This research focuses on the second
category of FDAs, given its high and ever-increasing adoption rate of urban
consumers (Q&Me, 2020).
2.2.3. Food delivery app market in Vietnam

Established in July, 2015 by Foody Corporation, Now had notched up notable
success, maintaining the market dominating position within the food delivery scene
for years. During mid-2016, the company claimed to have up to 10,000 orders per
day. Until 2018, the market for FDAs became hotter than ever, when new players
increasingly entered into the market, which included Lozi’s Loship, Ahamove’s Lala,
Vietnammm.com, Lixibook, together with such international rookies as Grab’s
GrabFood and GoJek’s GoFood. By October 2019, active players were reported to be
GrabFood, Now, GoFood, Baemin, Loship and Lixibook; and the market leader
position has switched to GrabFood, which is associated with its dominance in the
15
ride-hailing landscape. Specifically, this delivery giant reported a 250-fold growth in

orders together with thousands of restaurant partners after just one year in operation.
Statistics of Kantar TNS Vietnam show that, in 2018, GrabFood and Now have
"bet" their success on Vietnam’s online food delivery market by releasing countless
promotions with a view to converting users’ behavior towards food delivery.
Meanwhile, Lala by Ahamove suddenly announced their withdrawal after 01 year of
operating in December 2018. Besides, despite being one of the pioneering players in
Vietnam’s food delivery sector and financially backed by the Netherlands’
Takeaway.com, Vietnammm.com could not hold its position and ultimate accept an
acquisition deal with South Korea’s new unicorn Woowa Brothers Corporation,
which made inroads for Beamin to penetrate into Vietnam from June 2019.
The other two FDA players, Loship and Lixibook, have been quiet for months,
which reflects the toughness and intense competitiveness of this market. Since the
moment GoFood joined this lucrative market in November 2018 in HCM city and
March 2019 in Ha Noi, they have continuously released promotional campaigns to
attract customers and gain greater market share. In response to such competitive
moves from GoJek, Grab reinforces their promotion programs frantically to compete
with the Indonesian player. The intense race between these two FDA behemoths can
also be exhibited in the way they adopted Influencer marketing and invited famous
local celebrities to be their ambassadors. Conversely, former market leader, Now, so
far, has no intention to invest in huge marketing campaigns leveraging superstars.
Instead, it focuses on creating appealing discounts. Additionally, the South Korean
player, Baemin, despite stating that they have no plan to burn the process, still gave
away many promotion codes to raise customers’ awareness and obtain market share.
As of June 2020, based on Statista’s survey on 2,749 respondents, GrabFood

still occupies the position of market leader, becoming the most used apps for food
orders in Vietnam, which is followed by Now, GrabFood, Baemin and Loship,
respectively. This is consistent with survey findings of Q&Me by April 2020.
Moreover, similar to ride-sharing, the market for FDAs does focus on certain
growth drivers to ignite success, consisting of delivery speed, promotions, service
16
quality, market coverage, driver fleet, and the variety of merchant partners. Given
that promotions are a must for any business, and that delivery speed and the number
of driver partners are not subject to fierce competition, the FDA providers are
stepping up their game by scouting F&B merchant partners to diversify their menus.
Food delivery platforms have deployed various strategies to snag merchant

partners. For instance, many companies have adopted the solution of providing a
management platform, whose features include order tracking, menu and operation
hour updates, and even digital payments. As an early player within the market, Now
has long utilized this solution to assist its partners, in addition to promising
advertising merchant brands to millions of its users with its Foody ecosystem.
Meanwhile, Grab is taking advantage of its huge ride-hailing customer base to
promote its GrabFood services, in which the Singaporean unicorn leverages data-
driven analytical algorithms to derive suggestions for business optimization.
Aside from offering technology-related assistance, some food delivery

businesses also focus on crafting customized supporting solutions tailored to each
merchant partner, such as offering business consultancy or supporting food stall
owners in their efforts to establish a virtual store. Indeed, there exist several small
merchants who encounter limitations and challenges when doing business online.
When partnering with FDAs, they stand good chances of generating more income
and igniting their online success.
With surging demand for food delivery, combined with thorough support from
food delivery companies, several F&B merchants have enrolled into FDAs.
Specifically, there are over 20,000 stores providing food and drink together on Now
(Vietnam Credit, 2020). As regard GrabFood, it has expanded its network of more
than 8,500 F&B merchants during the short span of two years after its launch in 2018,
diversifying its options ranging from fast food to bubble tea and mala hotpot (Vulcan
Post, 2020). Statistics of D’ corp R-Keeper Vietnam and Statista show that Vietnam’s
F&B industry currently has around 540,000 eateries, including 22,000 cafeterias,
7,000 fast food restaurants, and over 80,000 facilities operated by large F&B chains.
If merely 3% of the nation’s restaurants go online, that would add another 19,000
17
merchants to delivery platforms, presenting a dramatic boom in the delivery market

(Vietnam Insider, 2020).
Seeing that not only FDA providers, but also F&B merchants, are interested
parties in this market, the study aims to deliver managerial implications for both.
2.2.4. Food delivery app usage in Vietnam

For a large number of people, especially Vietnamese young generation,
ordering food online has become a natural habit (Intage Vietnam, 2020). There has
been an extraordinary change in consumers’ lifestyles and living habits due to rising
urbanization and busy work schedules, which have obligated them to opt for
convenient food delivery options (B & Company, 2019; IMARC, 2019). The recent
survey on 600 FDA users by market research firm GCOMM indicated that up to 99
percent of respondents ordered food online 2-3 times a month. Another survey by
Q&Me on 840 respondents in HCM city and Hanoi in April 2020 (see Appendix 3)
also pointed out that more than 79% people order food at least 1 time per week while
around 7% order food every day and the other 6% order more than once a day.
Besides, the Southeast Asia report “Digital Consumers of Tomorrow, Here

Today" by Facebook and Bain & Company (2020) reveals that, while Vietnam’s e-
commerce growth is still highly fragmented, food delivery sees a less fragmented
landscape, in which one average Vietnamese consumers use only 02 food delivery
platforms. Also based on this report, top four factors driving Vietnamese FDA users’
satisfaction include delivery time, availability, product variety and brand assortment.
In comparison between HCM City and Hanoi - two most lucrative markets for
food delivery, the number of daily food orders in HCM City was six times higher
than that in Hanoi as of 2019, according to recent studies by Go-Jek and Kantar on
4,000 people aged from 15 to 45 in these cities. It is also shown that average spending
on food ordered online of the Saigonese is 10 percent higher than that of Hanoians.
In the light of COVID-19 pandemic and governmental orders for social

distancing in Vietnam, FDA adoption rate has been on a much more meteoric rise
(Facebook for Business, 2020; Q&Me, 2020). As the epidemic emerged and started
18
spreading alarmingly, a handful of FDA players has recorded a sudden surge of their
businesses – online orders have increased dramatically while a huge number of new
users have signed up and adopted the services regularly. A survey by GoFood reveals
that in the wake of the outbreak, even before the announced social distancing, from
February 2nd to 9th 2020, there were over 650,000 food orders on the platform, with
the figures increasing every day. The local startup Loship also witnessed record
business growth, with an 80% surge in the number of orders on its app in mid-March.
Based on Nielsen survey on the impacts of COVID-19 within Vietnam market,

the pandemic has resulted in over half of Vietnamese consumers reducing the
frequency of visiting physical stores while 82% reduced their out-of-home
consumption occasions, 45% increased their home food stocks and 34% ordered food
more from delivery services. Eating at home, either cooking at home or adopting
delivery services, becomes a common practice to avoid close contact when eating out
or buying take-away food, in which around 75% of survey respondents report to have
continued using or even use food delivery more often amid the outbreak. Such a shift
in eatery patterns has driven the prevalence of FDAs among Vietnamese consumers.
In fact, FDA remains top 5 digital mobile applications that have seen the
highest increase in first-time and continued usage during the first quarter of 2020
(Facebook and Bain & Company, 2020). According to Deloitte’s report (2019), FDAs
remain the biggest winner from the past crisis, with huge growing demand and
consumers who likely keep a higher level of usage after the outbreak. More than 94%
of FDA users claim they would keep using the platform(s) as often even after dining
establishments are open and social distancing measures are lifted (Statista, 2020).
2.3. The Millennials

2.3.1. Definitions of the Millennials
Millennials, also known as Generation Y, refer to the demographic cohort
following Generation X and preceding Generation Z. Members of this cohort include
individuals "reaching young adulthood in the early 21st century" (Oxford Living
Dictionaries, 2019). Yet, Millennials have been defined in various ways, each with a
different age range. Authors William Strauss and Neil Howe, well-known for creating
19
the Strauss–Howe generational theory and writing “Millennials Rising: The Next
Great Generation” (2000), are widely credited as ones who named the Millennials.
They defined the Millennials as one who were born between 1985 and 2000.
In a 2015 news release, the United States Census Bureau used the birth years
from 1982 to 2000 to describe Millennials while they have also stated that "there is
no official start and end date for when Millennials were born". Besides, the global
investment banking firm Goldman Sachs and the data and consulting company Kantar
Worldpanel define the Millennial generation as those whose birth years range from
1980 to 2000, which unites with the definition adopted by C Raines (2002), Hauw
and Vos (2010), Rainer and Rainer (2011), Hartman and McCambridge (2011), Lee
& Kotler (2016), etc. However, there have been several other definitions proposed by
academics and researchers. Some commonly used definitions are demonstrated in
Appendix 4.
Taking everything into consideration, the author decided to adopt the

definition of Millennials as those born between 1980 and 2000, because (1) this
definition is widely adopted by many academics and research firms, (2) this group
occupies large proportion of Vietnamese population (35% by 2018) (Kantar
Worldpanel, 2018) and (3) proves to be key drivers behind the profitability of FDAs
(Cheng, 2018; Suhartanto et al., 2019).
2.3.2. Some key characteristics of the Millennials

Sometimes termed as the “digital natives”, Millennial generation is generally
marked by elevated usage of and familiarity with Internet and mobile devices (Marc
Prensky, 2001). Living at the heart of the digital era, they are strongly influenced by
technology development. Thus, this demographic cohort has evolved distinctive
thought patterns, lifestyles and pre-purchasing expectations, which have greatly
differentiated their shopping behaviors from older generations (Kantar Worldpanel
Vietnam, 2018).
Being very tech-savvy consumers, Millennials nowadays leverage digital

platforms in every stage of their shopping journey to effortlessly access relevant
information and drive informed decisions. Shifting from passive traditional media to
20
more interactive channels, they are more willing to share opinions and leave reviews
for a particular product or service, as well as are easily influenced by the referrals and
reviews from others. Nonetheless, one drawback is that Millennials are becoming
more impatient, selective and skeptical of advertising or manufacturers' claims
(Kantar Worldpanel, 2018; Think with Google, 2018). Hence, the most effective
practice to win Millennials’ consumers is to deploy practical, authentic and emotional
communication.
What should also be noted is that this generational group has strong appetite
for new things. The majority of Vietnamese Millennials embrace innovative products
and concepts that answer their diverse needs and make their lives easier. Favoring
convenience and relevance, they are most responsive and willing to adopt new
shopping formats, such convenience stores and specialty stores and are also the
pioneer of online shopping (Nichols et al., 2015). Convenience associated with FDAs
is also the top driver shaping Millennials’ adoption and satisfaction (Alalwan, 2020;
Zhao and Bacao, 2020).
Furthermore, born between 1980 and 2000, several Millennials are at an age
where their personal and professional lives are just beginning to mature (Weber,
2020). While some of these young adults are settled down, many others just start their
career path, which explains why their purchasing power may be lower compared to
the older groups. Nonetheless, this does not mean Millennials are shopping cheaper,
but indeed, smarter. “Premiumization” trends amongst value-conscious Millennials,
even ones on a budget, are ongoing and will continue (McKinsey, 2017). They are
willing to pay premium price for products strongly associating with identities and
self-expressions (Valentine and Powers, 2013; Ordun, 2015). While Millennials
prefer e-coupons or promotions, a good deal is not always the lowest price, but the
best value for money, which is particularly true in case of FDAs (Smith, 2011; Palau-
Saumell et al, 2019).
When it comes to F&B industry, Millennials are considered “Intuitive

Foodies”, which means they enjoy eating and have a decided openness for trying new
flavors, even when they are anxious and concerned about finance. Millennials are
21
generally more engaged with food and drink, compared to other generations and their
intuitive choices are usually based on their focus on personal growth and development
(Mintel, 2019).
Seeing that Millennial accounts for around 35% of the population of Vietnam,
this generation remains main consumer target and thus still matters to many brands.
Particularly, this population is considered a key driver of FDAs’ profitability, being
the more regular users of these apps compared to Generation X and Baby boomers
(Cheng A., 2018; Chakraborty, 2019; My Nguyen, 2019; Nanaiah, 2020). The group
aged 25 to 34 years old, an important part of Vietnamese Millennials, accounts for
the largest share of FDA users (32.8% by October 2019) and also among those who
order food the most frequently via these platforms (Statista, 2019). Intage’s 2019
report on Vietnamese Youth Lifestyle also indicates that Millennials aged 20 – 30
have habitual patterns to use food delivery, especially at noon, both on weekdays and
weekends.
Not only to brand performance, Millennials, as the most avid digital users,
have shaped the media landscape and acquired the power to influence other
generations. Acknowledging that, marketers should enhance their understanding of
Millennial behavior to constructively engage and form long-term relationship with
them (Nyheim et al., 2015). For FDA providers and F&B merchants, they should dive
deep to gain crystal-clear insights into factors shaping these users’ satisfaction and
usage intention.
2.4. The concept of behavioral continuance intention to use

2.4.1. Behavioral intention to use
Although the concept of behavioral use intention has been used with varying
interpretations (Konerding, 1999), in numerous papers examining technology and
information system (IS) adoption, behavioral intention to use is based on Fishbein
and Ajzen’s (1975) definition as the strength of one person’s intention to engage in a
certain behavior. Intention is assumed to capture motivational factors that affect a
specified behavior, indicating how hard people are willing to try, how much of an
22
effort they are planning to exert, so as to perform that behavior (Ajzen and Fishbein,
1975, 1980).
In many prominent technology acceptance and use models, it has been either
suggested or indicated that intention to use the system is a reasonable indicator of
future system usage. These relationships are intended to capture “acceptance-like”
processes and deliver several managerial implications (Davis et al., 1989; Venkatesh
and Davis, 1996, 2000; Venkatesh, 2000; Venkatesh and Bala, 2008; Dwivedi et al.,
2011). Additionally, the research literature reveals that determination of factors
affecting behavioral intention to use, especially a technology-based IS, is very
significant to understand their factors’ roles in the successful implementation of
information systems.
2.4.2. Behavioral continuance intention to use (CI)

Continuance intention to use (CI) can be understood as ones’ intention to
continue using or long-term usage intention of a specific product/service
(Thominathan and Ramayah, 2015). As regards IS, CI refers to user’s decision to
continue to use a specific IS that he or she has already been using (Lee and Kwon,
2010).
Compared to initial intention to use, continuance intention has become a more

critical subject of study in IS research area (Lee and Kwon, 2010). Given that first-
time usage or initial acceptance of certain IS is a significant step toward realizing this
IS’s success, its long-term viability and eventual success depend on continued use
rather than first-time use (Bhattacherjee, 2001; Limayem et al., 2003). This can be
attributed to the fact that infrequent, inappropriate, and ineffective long-term use of
IS remains one of the main contributors to corporate failures (Lyytinen and
Hirschheim, 1987).
Continuance intention to use an IS, even at individual user level, is of

paramount importance to the survival of e-commerce companies (Bhattacherjee,
2001). Profitable and sustainable user base, market share, along with revenue streams
of these businesses depend on the number of both initial adopters and repeated
consumers. Yet, the overriding importance of continuance, vis-à-vis initial
23
acceptance, is evident from the fact that acquiring new customers may cost at least
five times in comparison with retaining existing ones (Parthasarathy and
Bhattacherjee, 1998; Baveja at al., 2000). Meanwhile, increasing consumer retention
rates by 5% grants business chances of growing their profits by 25% to 95% (F
Reichheld, 2001). Such figures speak volume about the importance of CI as a strong
indicator of success.
Notwithstanding that, continued usage only results in desired outcome if a

significant number of users move beyond the initial adoption and subsequently adopt
IS on a continued basis (Wangpipatwong et al., 2008). Furthermore, discontinuance
may occur after the adoption of innovation if the system does not meet the user’s
needs regardless of its successful prior adoption (Roger, 1995; Limayem et al., 2003).
Compared to the traditional economy, driving continuance usage and improving
consumer retention rate may be more challenging in the Internet economy.
Within such an ever-evolving digital era, today consumers, especially the

Millennials, become more demanding than ever before; they are more digitally
empowered to take their own informed decisions, and expect their needs to be met
instantly, perfectly, and tailored to their personality (Think with Google, 2018).
Besides, there exist a plethora of online and offline options for consumers to choose
from, and unless there are compelling reasons for choosing one particular company
over another, they tend to experiment or rotate purchasing decisions among multiple
companies (ET Crego and PD Schiffrin, 1995). Additionally, since the nature of
brand interaction with consumers has transformed dramatically, it can be expected
that key drivers of consumer’s continuance usage in the Internet economy may be
fundamentally different from that in the traditional economy. These facts underscore
the importance, relevance and timeliness of studying continuance as well as key
factors shaping consumers’ continuance intention as a topic of ample organizational
interest.
Being an information system itself, FDA(s), together with key factors shaping
continuance intention to use FDA(s), are also a critical subject for comprehensive
studies. Within the scope of this research, continuance use intention refers to
24
Millennial users’ intention to reuse, use continuously and/or use food delivery app(s)
in the long-term. In the light of blossoming delivery market and surging FDA usage
patterns in Vietnam, especially after the social distancing period amid COVD-19
outbreak, understanding determinants affecting FDA users’ continuance intention is
more essential than ever, particularly when it comes to FDA providers’ interest.
2.5. Conceptual models and theories on factors affecting continuance intention

to use food delivery apps
2.5.1. Theory of Reasoned Action (TRA) (1975)
Theory of Reasoned Action (TRA) was developed by Martin Fishbein and
Icek Ajzen (1975, 1980) with an aim to explain volitional behaviors (Bentler and
Speckart, 1979). TRA posits that individual’s intention to perform certain behavior
provides the strongest prediction of this behavior performance (Fishbein and Ajzen,
1975).
Behavioral intention is thought to be the result of Attitude toward the behavior

and Subjective Norm. Variables that are external to the model are assumed to
influence behavioral intentions only to the extent that they affect either attitudes or
subjective norms (Fishbein and Ajzen, 1975). TRA’s theoretical model, together with
variable definition, was exhibited in Appendix 5.
TRA proves to be an influential model for predicting the performance of

behavior and intentions, in which it has successfully applied to several contexts.
Nevertheless, one notable limitation of TRA lies in its predictive validity when the
behavior under study is not under complete volitional control. The prediction of
behavior from intention is also tough because there exist a variety of factors besides
one’s intentions which determine whether the behavior is performed (Sheppard et
al.,1988).
2.5.2. Technology Acceptance Model (TAM) (1989)

Whereas TRA generally aims to explain human behavior (Ajzen and Fishbein,
1980), Davis (1986) introduced an adaptation of TRA - the Technology Acceptance
Model (TAM) - which was specifically tailored to explain the intention to accept IS.
TAM adopts TRA as a theoretical basis for specifying causal linkages between two
25
key beliefs (perceived usefulness and perceived ease of use of an IS), and users'
attitudes, intentions and "actual computer adoption behavior". Similar to TRA, TAM
posits that actual system usage is determined by behavioral intention, but differs in
that subjective norm is not included and behavioral intention is regarded as being
jointly determined by user’s attitude and perceived usefulness (see Appendix 6).
Given that TAM is less generalizable than TRA, designed to apply only to IS
usage behavior, this model does incorporate several findings accumulated from over
a decade of IS research, hence, TAM may be especially well-suited for modeling user
acceptance for IS. Whereas several confirmatory results for TAM have been obtained,
there exist skepticisms among some researchers regarding the application and
theoretical accuracy (Davis at al.,1989; Brown at al., 2002; Yang and Yoo, 2004;
Bagozzi; 2007).
2.5.3. Task-Technology Fit (TTF) Theory (1995)

The task-technology fit (TTF) model by Goodhue and Thompson (1995) is a
well-established and widely-used theoretical model that underlines the significance
of task-technology fit in predicting how technology facilitate performance efficiency.
Figure 2.2: Task-Technology Fit (TTF) Theory (1995)
Source: Goodhue and Thompson, 1995

TTF was based on the proposition that both task characteristics and technology
characteristics can determine the task-technology fit, which in turn influences users'
performance and utilization. In other words, certain kinds of tasks require certain
kinds of technological functionality; if the gap between the task requirements and IS
functionalities widens, task-technology fit is reduced, which affects behavior of
employing the technology in completing tasks and individual performance adversely.
26
Since its initial proposal, this theory has gained much scholars’ interest and
extensively applied to a variety of IS (Larsen at al., 2009; Raven at al., 2010; Yen at
al., 2010; McGill at al., 2011; Aljukhadar at al., 2014; etc). Plus, the exploration of
connection between tasks and technology proves to be useful and warrants
continuance use of IS (Barki at al., 2007; Larsen at al., 2009; Lin; 2012; Wu and
Chen, 2017).
2.5.4. Unified Theory of Acceptance and Use of Technology (UTAUT) (2003)

Another prominent model adopted extensively in IS literature is Unified
Theory of Acceptance and Use of Technology (UTAUT) put forth by Venkatesh et
al. (2003). The model advances on the basis of synthesizing the dominant constructs
of eight prior prevailing models that range from human behavior to computer science.
As a framework specifically devised to explain technology acceptance in
organizational settings, UTAUT suggested four determinants of intention and usage
of information technology, which included Performance Expectancy, Effort
Expectancy, Social Influence and Facilitating Condition (see Appendix 7). Since its
original publication, the model has served as a baseline model for research on various
technologies in both organizational and non-organizational settings (Venkatesh et al.,
2012).
Notwithstanding that, a key element missing from UTAUT model is the

“individual” characteristics, which prove to be critical factors to predict technology
adoption and usage behaviors (Venkatesh et al., 2012; Dwivedi at al., 2019).
2.5.5. Unified Theory of Acceptance and Use of Technology 2 (UTAUT2) (2012)

In 2012, Venkatesh et al. proposed UTAUT2, which is based on the prior
UTAUT, yet tailored to individual consumer use context. In addition to four original
constructs, UTAUT2 incorporates three external constructs: Hedonic Motivation,
Price Value, Habit. Individual differences are also hypothesized to moderate the
effects of these seven constructs on behavioral intention and technology use.
27
Figure 2.3:Unified Theory of Acceptance and Use of Technology 2 (2012)
Source: Venkatesh et al., 2012

The extensions proposed in UTAUT2 produced a significant improvement in
UTAUT’s predictive validity within the consumer context. Given its recent
introduction in 2012, UTAUT2 is considered the most comprehensive theory in IS
research, with growing number of citations as a testimony to its popularity among IS
researchers, particularly for examining consumer-centered topics (Tamilmani at al.,
2017).
2.5.6. Expectancy Confirmation Model (ECM) (2001)

Expectation-Confirmation model (ECM) was proposed by Bhattacherjee
(2001), which is deemed one of the earliest theoretical models examining IS
continuance and taking into account the distinctions between acceptance and
continuance behaviors. ECM was conceptualized on the theoretical backdrop of
Expectation-Confirmation Theory (ECT) (Oliver, 1980), which investigated
consumer satisfaction and post-purchase intention (Oliver, 1980, 1981, 1993).
However, ECT has limitations in explaining IS expectation formation process since
it neglected IS quality factors and thus cannot thoroughly explain the notion of end-
user satisfaction (Khalifa & Liu, 2004).
28
Addressing ECT’s limitations, ECM posits that IS user’s intention for

continuance use is affected by three variables: user’s post-usage perceived usefulness,
the extent of user’s confirmation of expectations and his/her level of satisfaction with
IS. Nevertheless, post-usage perceived usefulness does not only affect users’
satisfaction, but also drives them directly to CI, since when they find a
product/service meets their needs, many will form a direct intention towards reusing
it without bothering to undergo the confirmation process. Likewise, confirmation
construct has an immediate effect on users’ satisfaction while also affecting their
post-usage perception of usefulness.
Figure 2.4: Expectancy Confirmation Model (ECM) (2001)
Source: Bhattacherjee, 2001
2.6. Empirical research on continuance intention to use food delivery apps

2.6.1. Foreign studies
Although FDAs have been well adopted worldwide, academic interest in
examining topics related to FDAs is still in its early stages (Ray et al., 2019; Wang et
al., 2019; Alalwan, 2020). Careful analysis reveals a number of empirical studies, but
only some of them cover the topic of CI related to FDAs. While such studies present
outstanding theoretical and practical contributions, there always exist some notable
gaps.
The research “Mobile food ordering apps (MFOA): An empirical study of the
factors affecting customer e-satisfaction and continued intention to reuse” by
Alalwan (2020) investigated key factors driving the e-satisfaction with MFOA and
customers’ behavioral intention to reuse such apps in Jordan. In this research, the
author employed an integrated model, which was based on UTAUT2 and the features
29
of MFOA (online review, online rating, and online tracking). Based on structural
equation modelling (SEM), the final statistical results supported the role of online
review, online rating, online tracking, performance expectancy, hedonic motivation,
and habit on e-satisfaction and continued intention to reuse, while failing to confirm
the impact of effort expectancy, facilitating conditions, social influence and price
value. Despite providing a theoretical contribution by adding three variables linked
to MFOA’s nature, this study was geographically restricted in Jordan and did not take
into consideration some important technical and cultural factors, such as mobile
interactivity, food habits or lifestyle.
Another research worth mentioning is the paper “What factors determining

customer continuingly using food delivery apps during 2019 novel coronavirus
pandemic period?” by Zhao and Bacao (2020) in China, which was carried out with
an aim of examining the antecedents of CI of FDAs during the COVID-19 pandemic.
This study proposed a comprehensive theoretical model, integrating UTAUT, ECM
and TTF with the factor of “trust”. Derived from 532 valid FDA users’ data, this
research findings revealed that satisfaction was the most significant factor, and
perceived task-technology fit, trust, performance expectancy, social influence and
confirmation have direct or indirect positive impacts on Chinses FDA users’ CI. Yet,
the associations between effort expectancy with other variables were not supported.
Besides, there were some limitations worth considering in this research. Firstly, the
scope of this study focused on FDA users in China and was conducted in a particular
context of COVID-19 pandemic, making it difficult to be generalizable to different
countries and contextual backgrounds. Besides, this research generally examined
customers' perspectives towards FDAs, without distinguishing the differences
between various genres of FDA platforms.
Also aiming to study the determinants of CI for FDAs, the research

“Determinants of Continuous Intention on Food Delivery Apps: Extending UTAUT2
with Information Quality”, conducted by Lee et al. (2019), adopted UTAUT2 that
augmented information quality. Once analyzing the survey answers from 340
respondents, the authors confirmed the role of information quality, performance
30
expectancy, habit, and social influence as factors driving FDA users’ CI, with habit
having the strongest influence. Regarding its gap, the research was conducted in
Korea, with cultural differences between single-person and multi-person households,
so generalizability of its findings may be limited.
Besides above-mentioned papers, there remain other relevant studies

conducted on intention to use or related topics, some of which are outlined in
Appendix 8.
2.6.2. Domestic studies

In Vietnam, so far, there have been just a few studies on online food delivery
service, especially almost no official research on consumers’ CI to use FDAs. One
relevant study to be referenced to is “How Food Delivery Services in Vietnam
Accommodate Millennials and Generation Z - Case company: Delivery Now”. This
research was conducted by My Nguyen (2019) with an aim of defining factors
affecting Millennials and Gen Z in their decision journey to choose a food delivery
service. Taking the inductive reasoning approach, this study suggested that
Vietnamese Millennials and Generation Z adopt food delivery services to satisfy their
need for convenience, which was why they first entered the decision-making journey.
For the following journey steps, they sought information through Google and social
medias (mainly Facebook and Instagram), highly trusted friends’ recommendations
that previously used food delivery services. While this research collected data from
interviews and surveys to explore factors driving customers in their purchasing
journey, it didn’t leverage proven theoretical frameworks or establish a new one to
investigate casual relationships among these factors. Besides, its data was collected
from only 178 responses, out of which only 135 responses fell into the target group.
In addition, due to the fact that this research focused solely on Delivery Now app, its
validity and generalizability can be a concern.
2.7. Hypothesis development and proposed research model

2.7.1. Rationale for the proposed research model
As already discussed in section (2.5.5), UTAUT2 is regarded as the most
comprehensive theory in understanding and predicting individual technology
31
adoption. The framework provided by UTAUT2 includes many critical constructs to

explain behavior within the consumer-focused context and has been extensively
adopted in several papers covering various fields, including some in food delivery
and FDAs (Venkatesh et al., 2012; Alalwan, 2020). Despite that, UTAUT2 focuses
more on users’ initial adoption, rather than their intention to reuse or/and use IS
continuously; thus, it should be extended or integrated with other theoretical model(s)
to better explain CI.
To address the gap of UTAUT2, the author incorporated it with ECM, which
has been widely implemented and also integrated with other adoption models to
investigate continuance intention of mobile technology (Zhao and Bacao, 2020).
Particularly, ECM has been integrated with TAM, TTF, UTAUT, UTAUT2,
Diffusion of innovations theory, etc. (Shang and Wu, 2017; Liébana-Cabanillas at al.,
2018, etc.) For instance, in the paper “Exploring the influential factors of continuance
intention to use mobile Apps: Extending the expectation confirmation model”, Tam
et al. (2020) deployed ECM and UTAUT2 to examine factors that underlie CI to use
mobile apps, and ultimately figured out that the combined framework enhanced the
predictive power in explaining CI.
Besides UTAUT2 and ECM, TFF was also integrated into the research model.
This model has been implemented by various previous studies to analyze users'
behavioral intentions of adopting mobile technology in different contexts, especially
in e-commerce. Compared to UTAUT2, TFF focuses more on the relationship
between task requirement and technology features, which proves to better explain
users’ technology acceptance (Goodhue and Thompson, 1995). TFF can also be
incorporated with other models like TAM, SEM, UTAUT, UTAUT2 (Zhou et al.,
2010; Yuan et al., 2016; etc.).
Therefore, it can be concluded that UTAUT2, ECM and the TTF have good
complementarities to evaluate the factors affecting behavioral continuance intention
to use FDAs, especially of Millennial users, who are tech-savvy and embrace
innovative products that answer their diverse needs (Kantar Worldpanel, 2018).
32
2.7.2. The proposed research model

Based on previous literature, this research model incorporated totally 11
variables from UTAUT2, ECM and the TTF model, with its hypothesis paths
presented as below:
Figure 2.5: The proposed research model
Source: Self-deprived by author, 2020

2.7.3. Hypothesis development
❖ Revisiting the UTAUT2 model
2.7.3.1. Performance Expectancy (PE)

Performance expectancy (PE) is defined as the degree to which a person
believes that IS adoption will allow for his/her better performance (Venkatesh et
al., 2003). This concept is in line with Perceived Usefulness in the TAM and ECM
model, and emerges as a key predictive factor for the behavioral intention of users.
As regards FDAs, PE refers to FDAs’ capability to enable their users to order and
receive food in a more convenient and productive way.
The role of PE has been repeatedly validated in numerous studies on the

factors determining the acceptance and use of new technologies and IS (Chong,
2012; San Martín, 2012; Venkatesh at al., 2012, 2013). Furthermore, PE has a
33
significantly positive effect on user's continuance usage of various mobile

technologies, such as mobile Internet (Zhou, 2011), mobile banking (Yuan et al.,
2016) and mobile shopping applications (Chopdar and Sivakumar, 2019).
Concretely related to food delivery and FDAs, PE proves to be positively correlate
with behavioral intention, in which, if users perceived higher utility from FDAs,
they are more likely to use and/or continue using them (Mun et al., 2017; Yeo et
al., 2017; Palau-Saumell at al., 2019; Roh and Park, 2019). Notably, in the study
“Mobile food ordering apps (MFOA): An empirical study of the factors affecting
customer e-satisfaction and continued intention to reuse” by Alalwan (2020), PE was
statistically validated to be the most influential to users’ continuance intention.
Hence, the first hypothesis (H) was proposed:
H1: Performance expectancy (PE) positively affects continuance intention (CI) to
use FDAs of the Millennials in HCM city.
Besides, PE has a significant effect on users’ satisfaction towards

(continuously) using mobile technology (Tam at al., 2020). Studies by Chong
(2013) and Marinković et al. (2020) provided statistical evidence supporting the
significant role of PE in contribution to users’ satisfaction towards mobile
commerce. Theoretically, ECM also posits that PE significantly influences the
satisfaction when adopting mobile technology. Thus, the following hypothesis
was suggested that:
H2: Performance expectancy (PE) positively affects satisfaction (SA) towards
continuously using FDAs of the Millennials in HCM city.
2.7.3.2. Effort Expectancy (EE)

As a fundamental variable of UTAUT and UTATUT2, Effort expectancy
(EE) can be simply understood as the extent to which using a new IS is easy and
requires little effort (Venkatesh et al., 2003). Considered a vital component in
technology acceptance, EE has been involved in several studies to examine CI of
information technology (Alagoz and Hekimoglu, 2012; Okumus and Bilgihan, 2014).
Whereas many researchers confirmed the positive correlation between EE and CI
34
(Kang, 2014; Fang and Fang, 2016), some studies rejected it (Yuan et al., 2016;
Chopdar and Sivakumar, 2019). Likewise, the impact of EE on FDA users’
continuance intention is also not consistent across empirical studies, with some
studies supporting (Ray et al., 2019) and the others not (Lee at al., 2019; Alalwan,
2020; Zhao and Bacao, 2020). Given such conflicting findings on EE’s influence and
geographical differences in the scope of these studies, the author still decided to keep
this and examine it within Vietnamese market.
H3: Effort expectancy (EE) positively affects continuance intention (CI) to use
FDAs of the Millennials in HCM city.
Meanwhile, EE has also been validated as a significant indirect predictor of

technology continuance usage, through its effect on other variables, such as PE
and satisfaction. Particularly, when examining users' CI of using mobile
technology, Yuan et al. (2016) suggested that EE significantly affected the PE.
Similar findings have also been affirmed in IS literature (Kim and Malhotra, 2005;
Shin et al., 2010). Likewise, many studies have provided further evidence
regarding the impact of EE on customer’s satisfaction towards continuance usage
of IS (Fang, 2016; Ray et al., 2019; Marinković, 2020; etc.). Hence, the two
following hypotheses were proposed:
H4: Effort expectancy (EE) positively affects performance expectancy (PE) of the
Millennial FDA users in HCM city.
H5: Effort expectancy (EE) positively affects satisfaction (SA) towards

2.7.3.3. Social Influence (SI)

Venkatesh et al. (2003) conceptualized social influence (SI) as the degree to
which one perceives that important others believe they should use the new system.
“Important others” refer to family members, friends, colleagues, influencers and any
people whose opinions, thoughts and attitudes are important to him/her. SI is viewed
as an important predictor of consumers’ (continuance) use intention, since they are
more likely to return to their social network either to acquire more information, or
35
to obtain social approval for their usage decision (Khalilzadeh at al., 2017;
Verkijika, 2018). Several papers demonstrated its significant role on users’
continuance intention to use mobile technology, including FDAs (Roh and Park,
2019; Suk Won Lee at al., 2019; Zhao and Bacao, 2020). Thus, the following
hypothesis was proposed:
H6: Social influence (SI) positively affects continuance intention (CI) to use FDAs
of the Millennials in HCM city.
Additionally, SI also indirectly formulates users’ intention to continuously

use mobile technology by affecting their satisfaction. Gallarza and Saura (2006)
suggested that social approval that consumers could have from others regarding
product/service adoption would enhance social values captured in using them, and,
accordingly, the level of consumer satisfaction. In line with this proposition, the
role of SI was also validated by Marinković et al. (2020) on mobile e-commerce
and Zhao and Bacao (2020) on FDAs. Hence, the following hypothesis can be
proposed:
H7: Social influence (SI) positively affects satisfaction (SA) towards continuously
using FDAs of the Millennials in HCM city.
2.7.3.4. Facilitating Condition (FC)

Facilitating Condition (FC) refers to the extent of availability of technical
infrastructure and human support which is perceived to help users facilitate their use
(Venkatesh et al., 2003). From a technical aspect, FDAs can be referred to as a mobile
application that could not be smoothly and effectively adopted by users without a
smartphone, Internet or 4G connection. Furthermore, many FDA users pay
considerable attention to the capacity of these platforms as well as their ability to
work consistently without downtime or technical problems. Besides, the role of
human support - in terms of customer service and call centers - is also important to
deliver high-quality service and largely shape users’ usage experiences (Alalwan,
2020). These explain why the variable of FC has been widely accepted by several
information technology and digital marketing researchers to exert a crucial impact on
36
the consumers’ behavioral intention and actual usage pattern. Concerning FDA users’
continuance intention, research findings by Lee et al. (2019), Alalwan (2020) and
Zhao and Bacao (2020) also statistically supported the impact of FC. Hence, the
following hypothesis was suggested:
H8: Facilitating condition (FC) positively affects continuance intention (CI) to

Given the critical role of FC underlying the Millennials’ CI to reuse FDAs, it

can also be proposed that this variable positively affects FDA users’ satisfaction.
From Alalwan’s perspective (2020), if users perceive an adequate level of technical,
organizational, infrastructural, and human support when using FDAs, they are likely
to undergo more comfortable experience of using these food apps; whereby, they tend
to derive greater satisfaction. Therefore, the author further proposed the below
hypothesis:
H9: Facilitating condition (FC) positively affects satisfaction (SA) towards

2.7.3.5. Hedonic Motivation (HM)

Together with such extrinsic motivations as performance expectancy (PE),
intrinsic motivation has repeatedly been observed as a strong driver of the
customer’s behavioral intention and willingness to adopt new systems and
applications (Davis at al., 1992; Brown and Venkatesh, 2005). As an important
intrinsic motivation, hedonic motivation (HM) can be understood as sense of
pleasure derived from adopting IS (Venkatesh et al., 2012), which could correlate
with the extent of innovativeness and novelty of those systems (Van der Heijden,
2004).
Likewise, since such technology-based applications as FDAs, are perceived

as being modern and innovative, consumers may have feelings of enjoyment when
using them (Okumus and Bilgihan, 2014; Yeo et al., 2017; Okumus et al., 2018).
37
Specifically concerning FDAs, Alalwan (2020) also validated the significance of

HM in users’ continuance use. Therefore, the following hypothesis was suggested:
H10: Hedonic Motivation (HM) positively affects continuance intention (CI) to

Additionally, many FDA providers empower users to co-create value by

providing feedback in online reviews and ratings of the delivery services and the
meal itself (for instance, features of online rating for delivery drivers offered by
Now, GrabFood, GoFood and Baemin). Thanks to that, FDA users tend to
perceive their important role both for other customers and for service providers,
which may enhance their feeling of pleasure. Given the role of HM as discussed
above, customers could obtain more satisfaction with their usage experience if
they perceive intrinsic motivation in using such apps. The following hypothesis
was proposed accordingly:
H11: Hedonic Motivation (HM) positively affects satisfaction (SA) towards

2.7.3.6. Habit (HB)

As the last construct of UTAUT2 model, habit (HB) is defined as the extent
to which people incline to automatically engage in certain behaviors because of
learning (Limayem et al., 2007). Within this ever-evolving mobile landscape,
individuals are increasingly attached to their smartphones and form a habitual
behavior towards using associated mobile apps (Shareef et al., 2012, 2016). The
outcomes of such accumulated learning experience could drive consumers’
attitudes and beliefs, which then affect their CI to behave in the same manner as
previously (Ajzen at el., 2005).
Empirical studies have clearly validated the pivotal role of HB in the area
of mobile commerce and mobile app adoption/continuance intention (Morosan
and DeFranco 2016, Amoroso and Lim, 2017; Rana et al., 2017; etc). The research
“User Acceptance of Mobile Apps for Restaurants: An Expanded and Extended
38
UTAUT-2” by Palau-Saumell et al. (2019) also indicated that habit was the
strongest predictor of intentions to use, and of actual usage, in the context of
mobile apps for restaurants. Aligning with these findings and Venkatesh et al.’s
(2012) proposition, the Millennial FDA who already form a habitual behavior
towards FDAs will intend to keep using these apps in future. For this reason, this
paper suggested another hypothesis:
H12: Habit (HB) positively affects continuance intention (CI) to use FDAs of the
Millennials in HCM city.
2.7.3.7. Price-saving Orientation (PO)

The construct of price value is a theoretical addition to UTAUT-2 by
Venkatesh et al. (2012) as individuals bear monetary cost in using technology in
consumer setting. However, this construct is not applicable in the models when the
technologies examined, such as FDAs, were available to users free of cost (Palau-
Saumell et al., 2019; Tamilmani et al., 2019). Several studies also excluded this
variable from their research model when explaining technologies such as mobile
applications and social networking sites that are available to individuals without
incurring any financial cost (Koenig-Lewis et al., 2015; An et al., 2016; Tamam et
al., 2016; etc.). Thus, the author decided not to include the variable of price value
into the research model.
Instead, the construct of price-saving orientation (PO) was suggested to

incorporate into the model. PO can be understood as the economic values to be
obtained by technology users, which means that the use of technologies allows users
to acquire a product or service at a lower or more reasonable price (Jensen, 2012). In
several previous studies on UTAUT2, price value construct has been replaced with
PO and these authors statistically validated a positive relationship between PO and
behavioral intentions (Lee and Joshi, 2007; Escobar-Rodríguez et al., 2013; Escobar-
Rodríguez, 2014; etc).
Indeed, most consumers, especially price-conscious ones as the Millennials

(Kantar Worldpanel, 2018), actively look for price discounts as they are concerned
39
over the amount of money that they are able to save through these discounts
(Darke et al., 1995). The power of price discounts is also supported by evidence a
research stating that discounts will increase the perceived value to the offer of
certain product/service and stimulate sales for businesses because it indicates that
the price is an even better bargain (Madan and Suri, 2001; Thaler, 2008).
Within the F&B sector, different segments of consumers are likely to opt
for different levels of food quality and price. Given that, almost all consumers tend
to rationalize and make value-driven decisions based on the maximum benefit that
they can get out of that deal by looking for the lowest acceptable price (Ollila,
2011). Regarding online food landscape, the Internet empowers consumers to
compare prices among different providers, whereby enabling them to purchase
meals at a more reasonable cost (Gentry and Calantone, 2002). In FDAs, the
availability of discounts and special offers, together with information on products
and prices, will allow for price saving, which, in turn, drive users’ satisfaction and
continuance intention to use (Palau-Saumell et al., 2019). Thus, the following
hypotheses were proposed:
H13: Price-saving orientation (PO) positively affects continuance intention (CI)
to use FDAs of the Millennials in HCM city.
H14: Price-saving orientation (PO) positively affects satisfaction (SA) towards

❖ Revisiting the TTF theory
2.7.2.8 Perceived task-technology fit (TTF)

Goodhue and Thompson (1995) conceptualized Perceived task-technology fit
(TTF) as the extent to which an IS supports individuals in accomplishing their
portfolio of tasks. This construct aims to measure the match among a user’s
requirements for a specific task, the user’s abilities and the functionality of a certain
technology. Within this research’s scope, TTF represents the advantages of FDAs
that allow users to conveniently order food and manage delivery process to ensure
the quality of the service.
40
As a crucial factor summarized from the task-technology fit model, TTF

performed a significant role in formulating users’ PE in technology
adoption. Incorporating TFF model with UTAUT, both Zhou et al.
(2010) and Oliveira et al. (2014) found out that the construct of TTF was a strong
determinant of PE in terms of mobile banking adoption. Similar findings have also
been empirically proven on the field of mobile e-commerce (Yuan et al., 2016) and
FDAs (Zhao and Bacao, 2020). For the reason stated, this study assumed the
following hypothesis:
H15: Perceived task-technology fit (TTF) positively affects performance
expectancy (PE) towards continuously using FDAs of the Millennials in HCM city.
❖ Revisiting the ECM model
2.7.3.9 Confirmation (COF)

Confirmation (COF) of expectations refers to the extent to which one’s
perception over an IS is compatible with his prior expectations and its actual
performance (Bhattacherjee, 2001). Within ECM model, Bhattacherjee (2001)
theorized COF as an influential predictor of PE and satisfaction towards IS usage
since it implies the realization of expected benefits of IS use. Literature review
suggested several research evidences supporting the significant role of COF in
formulating PE and sense of satisfaction of IS users. In the paper “Exploring the
influential factors of continuance intention to use mobile Apps: Extending the
expectation confirmation model”, Tam et al. (2018) validated the strong relationships
between COF and PE and between COF and satisfaction with mobile apps. Findings
of many other ECM-based studies, on a plethora of mobile technologies' continuance
usage contexts, have also been consistent with the given results. Hence, the following
hypotheses were proposed in this study:
H16: Confirmation (COF) positively affects satisfaction (SA) towards
H17: Confirmation (COF) positively affects performance expectancy (PE)

towards continuously using FDAs of the Millennials in HCM city.
41
2.7.3.10. Satisfaction (SA)

Within the consumption context, satisfaction (SA) was defined by Oliver
(1981) as an accumulated psychological state when the emotion surrounding
disconfirmed expectations is coupled with his/her prior feelings about usage
experience. According to ECM model examining IS continuance, SA refers to an
overall emotion-based evaluation of an IS (Bhattacherjee, 2001). ECM supports the
belief that SA towards using a product or service is the primary motivation for its
continuance, stating that the direct relationship between SA and CI lies at the core of
the IS continuance model.
A myriad of empirical papers in various fields of IS and mobile technologies
have also examined and confirmed such a strong relationship (Dlodlo, 2014; Cao et
al., 2018; Gao et al., 2015; Susanto et al., 2016, etc). In terms of FDA usage, Alalwan
(2020) and Zhao and Bacao (2020) validated the role of SA in driving users’ intention
of reusing FDAs, with SA being one of the strongest predictors of FDA users’ CI.
For the reasons stated, SA, as a complementary variable of UTAUT2 and ECM, has
been proposed in the following hypothesis:
H18: Satisfaction (SA) positively affects continuance intention (CI) to use FDAs
of the Millennials in HCM city.
In this chapter, the author presented three (03) content pillars. Firstly, an
overview of food delivery, food delivery apps and Millennials’ distinctive
characteristics was demonstrated, in which market performance and consumers’
usage patterns are highlighted, both in global and national context, particularly within
HCM city. Secondly, the author exhibited literature review of key conceptual models
and theories on behavioral (continuance) intentions towards IS as well as some
outstanding studies on CI to use FDAs. Some basic concepts of behavioral intention
and continuance usage intention were also explained. Thirdly, the author proposed
the research hypotheses and model, along with the definition and rationale for each
variable within the model. These were all derived from literature review of relevant
theories and empirical studies.
42
CHAPTER 3: RESEARCH METHODOLOGY

3.1. Research process
The detailed research process consisted of two main phases, namely
preliminary research and official research, which was demonstrated in Figure 3.1 as
follows:
Figure 3.1: Research process
Research topic
Research aims and objectives, subject

and scope
Preliminary research
Literature review 1st draft of questionnaire
Pretest 2nd draft of questionnaire

(Expert review & focus group)
pretest
Pilot study (n = 40) Official research questionnaire
Cronbach's Alpha
Official quantitative research (n = 457)

- Cronbach’s Alpha coefficient analysis
- Explanatory Factor Analysis (EFA)
Official research
- Confirmatory Factor Analysis (CFA)

- Structural Equation Modeling (SEM)
- T-test & ANOVA analysis
Result analysis and discussion
Recommendations
Source: Self-deprived by the author, 2020

43
3.2. Preliminary research

3.2.1. Initial questionnaire development
3.2.1.1. Measurement scale construction based on theoretical basis
As all 11 constructs within this model (namely, PE, EE, SI, FC, HM, HB, PO,
TTF, COF, SA, CI) are social-economic concepts that have many facets, the author
had to leverage multivariate measurements to obtain more accurate result (Hair et al,
2014).
To do so, the author constructed scale for each of these variables through
synthesizing measurement items from well-established research papers, including
ones by Goodhue and Thompson (1995), Bhattacherjee (2001) and Venkatesh et al.
(2003, 2012). Specifically, all measurement items, together with its Cronbach's Alpha
coefficient (CA) for measurement reliability, were depicted in the table below:
Table 3.1: Measurement scales for all variables
Item Statement CA Source

Performance Expectancy (PE)
I feel that food delivery apps (FDAs) are useful for ordering and 0.881
PE1
receiving delivery food Bhattacherjee
(Zhao and
PE2 I feel FDAs are convenient to order and receive delivery food (2001), Venkatesh
Bacao, 2020)
Using FDAs improves the process of ordering and receiving delivery et al. (2003), (Zhao
PE3
food and Bacao, 2020)
Using FDAs improves the efficiency of ordering and receiving
PE4
delivery food
Effort Expectancy (EE)
EE1 Learning how to use FDAs is easy 0.883 (Zhao Venkatesh et al.
EE2 Interaction with FDAs is clear and comprehensible and Bacao, (2003), Yuan et al.
EE3 It is easy to follow all the steps of FDAs 2020) (2016), (Zhao and
EE4 It is easy to become skillful at using FDAs Bacao, 2020)
Social Influence (SI)
SI1 0.860 (Zhao Venkatesh et al.
People who are important to me recommend that I use FDAs
and Bacao, (2003), (Zhao and
SI2 People who are important to me think FDAs are beneficial 2020) Bacao, 2020)
SI3 People who are important to me think it is a good idea to use FDAs
SI4 People who are important to me support me to use FDAs
Facilitating Conditions (FC)
FC1 I believe that I have the necessary smartphone to use FDAs 0.83 (Palau- Venkatesh et al.
FC2 I believe that I have the necessary knowledge to use FDAs Saumell et (2012), Palau-
FC3 I believe FDAs are compatible with other technologies I use al., 2019) Saumell et al.
FC4 I feel comfortable using FDAs (2019)
Hedonic Motivation (HM)
HM1 I believe that using FDAs is fun
44
HM2 I believe that using FDAs is enjoyable 0.916 Won Venkatesh et al.
HM3 Lee at al. (2012), Won Lee at
I believe that using FDAs is very entertaining (2019) al. (2019)
Price-Saving Orientation (PO)
PO1 I can save money by examining the prices of different restaurants 0.82 Palau- Jensen (2012), Wen
when using FDAs Saumell et (2012), Palau-
PO2 I like to search for cheap restaurant deals when using FDAs al. (2019) Saumell et al.
PO3 I believe FDAs offer better value for my money (something that is well (2019)
worth the money spent on it)
Habit (HB)
HB1 The use of FDAs has become a habit for me 0.889 Won Venkatesh et al.
HB2 I am in favor of using FDAs Lee at al. (2012), Won Lee at
HB3 I feel the need to use FDAs (2019) al. (2019)
HB4 Using FDAs on my smartphone has become natural to me
Perceived task-technology fit (TTF)
TTF1 The functions of FDAs are enough for me to order and receive the 0.880 (Zhao Goodhue and
delivery food and Bacao, Thompson (1995),
TTF2 The functions of FDAs are appropriate to help manage the ordering 2020) (Zhao and Bacao,
and receiving the delivery food 2020)
TTF3 The functions of FDAs fully meet my requirements of ordering and
receiving the delivery food
Confirmation (COF)
COF1 My experience with using FDAs is better than what I expected 0.848 (Zhao Bhattacherjee
COF2 The functions of FDAs are more than what I expected and Bacao, (2001), (Zhao and
COF3 The service provided by FDAs is better than what I expected 2020) Bacao, 2020)
COF4 Overall, most of my expectations from using FDAs were confirmed
Satisfaction (SA)
SA1 I am very satisfied that FDAs meet my requirements 0.848 (Zhao Bhattacherjee
SA2 I am satisfied with FDAs efficiency and Bacao, (2001), Hung et al.
SA3 My interaction with the FDAs is very satisfying 2020) (2012)
SA4 I think I did the right thing by using FDAs
Continuance Intention (CI)
CI1 I intend to continue using FDAs in the future 0.916 (A.A. Bhattacherjee
CI2 I will always try to use FDAs in my daily life Alalwan, (2001), Venkatesh
CI3 I plan to continue to use FDAs frequently 2020) et al. (2012)
CI4 I have decided to use FDAs for purchasing foods the next time
3.2.1.2. Questionnaire design

The questionnaire was designed with 54 questions, with the target respondents
aged from 20 to 40 (Millennials, born between 1980 and 2000), who use FDAs and
are currently living in HCM City. As the almost all respondents were expected to be
Vietnamese, the questionnaire was written in Vietnamese, with all above
measurement scales being translated and adjusted to be as natural as possible to the
Vietnamese language. Concerning the questionnaire’s structure, there were three
main parts:
45
Part I consisted of 6 questions, aiming to collect general data on FDA usage

patterns of respondents, particularly on their usage frequency and favorites apps.
Part II was the most important part of the questionnaire as there presented all
key questions whose responses would be garnered for quantitative analyses to
determine factors formulating FDA users’ CI. This part included 41 questions in form
of statements and adopted Likert scales. The scale was assigned value from 1 to 5,
which respectively referred to "Completely disagree", "Disagree", "Neutral",
"Agree", " Completely agree".
Part III covered 7 questions on basic demographic information of respondents

along with open questions on users’ recommendation to improve FDA and its
service(s). This purpose of those demographic questions was to filter the
demographic information of respondents, allowing the author to classify and ascertain
that collected samples relatively reflected demographic situation of HCM City.
3.2.2. Pretest: Expert reviews and focus groups

According to Nguyen Dinh Tho (2012), a good questionnaire has to go through
many review and trial rounds and is adjusted properly to ensure its efficiency for
official research. Thus, the author decided to leverage some pretest techniques,
particularly expert review and focus group interview, to evaluate the questionnaire’s
first draft.
3.2.2.1. Expert review

The author consulted 04 experts within the Marketing and Communication
fields, as specified in Appendix 9. Generally, all these experts had no but minor
remarks on the research questionnaires. They suggested revising how some questions
were worded to be natural to Vietnamese language as well as adding some fine details
to explain questions better and avoid confusion. Besides, one expert proposed adding
the word(s) indicating continuance usage intention into some measurement statement
(for instance, the statement “People who are important to me recommend that I use
FDAs” was revised into “People who are important to me recommend that I
use/continue using FDAs”). Except these feedbacks, there was no suggestion for
omitting or adding more questions.
46
3.2.2.2. Focus group interview

After adjusting the questionnaire, the author performed focus group
interviews, on October 19th and 20th 2020, within three small groups: Group 1
included 05 University students, group 2 included 04 first jobbers and group 3
included 03 white-collar experienced employees. A noteworthy point is that all
interviewees belonged to different age groups, all being tech-savvy and using FDAs.
These interviews were conducted through a pre-designed outline (Appendix

10), together with the revised questionnaire, and under the author’s guidance.
Notably, data gathered from this step did not intend for data processing and statistical
analysis, but only used to assess questionnaire quality. While interviewees were
responding to the questionnaire, the author followed the entire process and observed
their interactions and facial expressions (especially, grimaces or gestures indicating
confusion or hesitation).
As regards the interview outcomes (Appendix 11), almost all interviewees

confirmed the impact of factors proposed in research model on their continuance
intention to use FDAs. Concerning specific measurement statements for each factor,
interviewees agreed that they were well defined, clearly understood and presented in
a consistent manner. Meanwhile, one person suggested abbreviating Vietnamese
noun phrase of FDAs (i.e. the phrase “(các) ứng dụng giao thức ăn trực tuyến” should
be abbreviated into “FDA(s)”) to make all statements more concise. However, many
others opposed this view, saying that the abbreviation may result in confusion,
especially when FDA has already abbreviated for other phrases, such as the Food &
Drug Administration.
After performing these above pretest techniques and considering the

adequacy of all comments, the author decided to revise the questionnaire in
accordance with all revisions proposed by experts and still keep the phrase “(các)
ứng dụng giao thức ăn trực tuyến” to minimize any chance of confusion, from
which the second draft of research questionnaire was formed and prepared for pilot
study.
47
3.2.3. Pilot study

Once the second questionnaire draft had been completed, the author conduct
pilot study to verify its understandability and efficiency. Regrading sample size, Fink
(2003) validated that the minimum number for pilot study was 10 while Browne
(1995) stated this should be at least 30. Given that, the author conducted a pilot test
on a small sample size (n=40) and evaluated the reliability of all measurement
instruments using Cronbach’s Alpha coefficients. Among 40 respondents, there were
29 University students and 11 office staffs, all being Millennials, living in HCM city
and using FDAs.
In regards the result of Cronbach’s Alpha test on pilot study data, all
measurement scales were validated to be reliable upon the first test run, thereby being
able to be leveraged for designing official questionnaire and performing official
research (full results of Cronbach’s Alpha analysis were shown in Appendix 12).
3.2.4. Official questionnaire

Once pretest and pilot study were performed, the questionnaire was revised
accordingly and its final version was crafted. The finalized measurement scale and
questionnaire were exhibited in Appendix 13 and Appendix 14 respectively.
Leveraging Likert Scale, the questionnaire focused on examining respondents’
continuance intention to use FDAs as well as factors formulating such intention.
Since there was no question added or omitted, there remained 54 questions, including
41 quantitative questions.
3.3. Official research

3.3.1. Data collection and sample size determination
3.3.1.1. Data collection
This research utilized convenience sampling, which referred to a non-
probability sampling type where members of the target population meeting certain
practical criteria are included (Etikan et al., 2016). In particular, a questionnaire
survey was designed and distributed online via social networks and email to eligible
individuals. Most responses were accepted except ones by those living outside HCM
city, never using FDAs and not being Millennials themselves. Specifically, out of 472
48
responses received, there were 457 valid responses which would then be used as input
for statistical analysis.
3.3.1.2. Sample size determination

Data analysis methods used in this study included CFA and SEM. Although
most researchers and scholars have agreed that SEM requires large sample sizes, there
is no single correct or universally-accepted calculation or method for determining
sample size for SEM. Whilst Hoyle (1999) and Hoyle and Kenny (1999) suggested
that SEM could be significantly adopted even if sample size is quite small, N = 100 -
150 is considered the minimum sample size for conducting this modelling method
(Tinsley and Tinsley, 1987). Although these figures can be referenced, sample size
determination should be model-specific and be subject to research context (Westland,
2010; Wolf et al., 2013). Besides, in terms of EFA analysis, Hair et al. (1998)
validated that the number of observed variables in the survey would determine sample
size n, in which n should be at least 5 times of observed variables. By doing that, the
smallest sample size for this study would be 5x41=251. All things considered, n=457,
as the number of valid responses collected from survey, should constitute an
appropriate sample size for this paper.
3.3.2. Data analysis

3.3.2.1. Descriptive statistics
Once data was collected, descriptive statistics was applied to describe the
characteristics of the samples while also detecting any unusual values in the collected
samples (Pallant, 2007). Measures that are mainly used from descriptive statistics
include the mean, minimum, maximum together with the standard deviation of each
of the variables. Besides, the output of this section for demographic data is a
descriptive profile of the sample, which was illustrated through Gender, Monthly
income, Age and Educational background.
3.3.2.2. Cronbach’s Alpha coefficients

Among the most crucial qualities of academic research, reliability is defined
as the degree to which the measurement scale is regarded as consistent and stable
(Parasuraman et al., 1991). One of the most popular methods leveraged to evaluate
49
the scale reliability in a questionnaire survey is Cronbach's Alpha coefficients

(Cronbach, 1951), which allows researchers to identify which items should be
retained as well as which should be eliminated out of each measurement. Specifically,
the reliability index of the measurement scale is assessed through two factors, which
include:
(1) Cronbach’s Alpha coefficient: Its value ranges from zero to one, in which
higher values implies greater internal consistency within a group of items (Sekaran,
2003). Table 3.1 manifested the ranges of Cronbach's Alpha value together the
adequacy level of their internal consistency as specified by DeVellis (1991).
Table 3.2: Adequacy of Cronbach's Alpha value

Cronbach's Alpha Adequacy Cronbach's Alpha value Adequacy
value
Above 0.95 Redundant 0.65 – 0.70 Minimally

Acceptable
0.80 – 0.95 Very Good 0.60 – 0.65 Undesirable
0.70 – 0.80 Respectable Below 0.60 Unacceptable
Source: DeVellis, 1991
(2) Corrected Item — Total Correlation: The value of each measurement items
should display the figure equal or higher than 0.30 to qualify the test; otherwise, such
items should be removed from the scale (Nunnally, 1978; Pallant, 2007).
3.3.2.3. Exploratory Factor Analysis (EFA)

While Cronbach's Alpha coefficient aims to examine the connection of a set
of items within a variable, Exploratory Factor Analysis (EFA) aims to examine the
degree of correlation among different variables, in which the number of constructs
underlying each variable could be controlled to establish a more manageable set
(Pallant, 2007). Since this study adopted CFA and SEM, there were four specific
noteworthy indicators:
(1) Kaiser – Meyer – Olkin (KMO) coefficient: This refers to the adequacy of factor
analysis. KMO coefficient within the range 0.5 and 1.0 (0.5 ≤ KMO ≤1.0) implies
that factor analysis is applicable for the data set.
50
(2) Bartlett's test of sphericity: This indicator aims to evaluate whether certain
measurement items in a factor are correlated or not. According to Nguyen Dinh Tho
(2011), Bartlett test must have significance level lower than 0.05 (Sig < 0.05) so that
"convergence validity" of items in each measurement scale can be established.
(3) Total Variance Explained: This indicator aims to examine the suitability of EFA
analysis, which should be at least 50% (Total Variance Explained ≥ 50%).
(4) Factor loading: This indicator manifests the correlation relationship between
observed variables and factors. A larger Factor Loading value means the better
correlation. The minimum acceptable value of this indicator is 0.5.
3.3.2.4. Confirmatory Factor Analysis (CFA)

The next step after EFA is Confirmatory Factor Analysis (CFA), which proves
to be a critical analysis tool in a myriad of fields in social and behavioral sciences
(Mueller and Hancock, 2001). While EFA aims to explore the factor structure, CFA
helps to confirm the factor structure already extracted in the EFA. This statistical
technique offers a measurement model based on SEM, aiming to assess the fit
between observed data and the theoretically grounded model. Specific benchmarks
for model fit, reliability and validity of CFA model were illustrated at Table 3.3.
3.3.2.5. Structural Equation Modeling (SEM)

Structural equation modeling (SEM) has been validated as one in-depth
statistical approach to examine hypotheses about relationships among observed and
latent constructs (Hoyle, 1995). Compared to other research methodologies, SEM is
more compatible for studies with comprehensive theoretical frameworks in which
influence of constructs are propagated across various layers of variables
(Baumgartner and Homburg, 1996). Since this study leveraged a research model,
which integrated UTAUT2, TFF and ECM as well as hypothesized many layers of
direct and indirect relationships, the author decided to adopt SEM, which is deemed
more effective than commonly-used multiple linear regression (Cheng, 2001). In term
of model fit, reliability and validity indices, its assessment criteria are comparable to
ones for CFA (Table 3.3).
51
Table 3.3: Criteria for assessment CFA/SEM statistical results
Criterion Index Abbreviation Standard
Reliability Standardized Loading Estimates Estimates ≥ 0.5: Acceptable

Estimates (Factor Loading) Estimates ≥ 0.7: Ideal
Composite Reliability CR CR ≥ 0.7
Convergent Average Variance Extracted AVE AVE ≥ 0.5

validity
Discriminant Maximum Shared Variance (MSV) < Average Variance Extracted (AVE)
validity
Square Root of AVE (SQRTAVE) > Inter-Construct Correlations
Model fit Chi-square (x2) CMIN p-value < 0.05
Chi-squared adjusted for CMIN/df CMIN/df ≤ 2: Good

degrees of freedom CMIN/df ≤ 5: Acceptable
Goodness of fit index GFI GFI ≥ 0.8: Acceptable

GFI ≥ 0.9: Good
GFI ≥ 0.95: Very good
Comparative Fit Index CFI CFI ≥ 0.9: Good

CFI ≥ 0.95: Very good
Tucker-Lewis Index TLI TLI ≥ 0.9
Root Mean Square Errors of RMSEA RMSEA≤0.08: Good

Approximation RMSEA≤0.03: Very good
P-value for a test of close fit PCLOSE PCLOSE ≥ 0.01: Acceptable

PCLOSE ≥ 0.05: Good
Source: Hu and Bentler, 1999; Hair et al., 2010
3.3.2.6. Independent Sample t-Test & One-Way ANOVA

After structural modelling had been conducted, the author performed an
examination over the impacts of demographic observations on the key dependent
variable, Millennials’ continuance intention to use FDA, through Independent
Samples t-Test and one-way ANOVA. Independent Samples t-Test is intended to
compare the means of two independent groups so as to identify if there is statistical
evidence that the associated population means are significantly different (Rochon and
Kieser, 2012). Accordingly, the author conducted this test on demographic variables
of Gender to evaluate whether differences in this demographic characteristic result in
any differences in users’ continuance intention. As regards the testing result,
significance value of Levene's Test and t-Test at “Equal variances not assumed”/
52
“Equal variances assumed” would be utilized for evaluation. Because t-Test cannot
be utilized to make comparisons among more than two groups, one-way ANOVA
would be used instead, particularly when it comes to Monthly income, Age and
Educational background. In the ANOVA analysis, the author also carried out the
Levene’s Test to verify the difference in demographics variance. Next steps of one-
way ANOVA analysis would be briefly described as follows:
Figure 3.2: Procedures for testing one-way ANOVA
Sig value for

Levene’s Test
Sig ≥ 0.05 Sig < 0.05
ANOVA Robust Test

table table
Sig for F-test Sig for F-test ≥ Sig for Welch Sig for Welch ≥
<0.05 0.05 <0.05 0.05
No statistically Statistically No statistically

Statistically significant significant significant difference
significant difference difference difference
Source: Samuel et al., 2005
This chapter provided the whole methodology to be applied in the research,
ranging from general research process to each detailed research step. For preliminary
research, the author adopted two pretest techniques, including expert review and
focus group interview, and a pilot study on small sample size (n=40), in order to test
the compatibility of measurement scales and perfect the final questionnaire for
official research. Regarding measurement scales’ reliability, statistical results after
running Cronbach’s Alpha coefficients on pilot study data proved to be satisfactory;
therefore, the latter phase should be proceeded. For official research phase,
questionnaire survey was conducted to collect larger-scale sample (n=457) that then
underwent six analytical steps (Descriptive statistics, Cronbach’s Alpha, EFA, CFA,
SEM, t-Test/ANOVA) to validate the significance of each variable and test the
proposed hypotheses.
53
CHAPTER 4: RESEARCH RESULTS

4.1. Descriptive statistics analysis
4.1.1. General description
After distributing the questionnaire surveys via social networks and through
emails, the author collected 472 samples. After preliminary check, a total number of
457 responses were kept for data analysis while the other 15 samples were excluded
since they did not match respondents’ requirements. The sample size of 457 was
considered to be compatible for research model running EFA, CFA and SEM (Tinsley
and Tinsley, 1987; Anderson and Gerbing, 1988; Ding et al.,1995; etc). As for the
process, garnered data was first coded and imported into Microsoft Excel 2019 for
cleaning purpose before being proceeded into IBM SPSS Statistics 20.0 and IBM
SPSS AMOS 24.0.
4.1.2. Statistical description

4.1.2.1. Descriptive statistics analysis by demographics factors
Demographic factors included in this research were Gender, Age, Monthly
income, Educational background and Current living location; yet, since valid
responses all lived in HCM city, only 4 first given demographic factors were imported
for descriptive statistics analysis. Full descriptive statistics of these demographic
factors, together with its summary of key findings synthesized by the author, were
exhibited in Appendix 15.
In terms of Gender, the majority of responses are from female respondents,

which represents 68.3% compared to 31.7% of male. This is consistent with survey
by Statista (2020) and could be explained by the fact that Vietnamese female usually
take slightly more interest in ordering food online than male counterparts (Kim Dang
at el., 2018).
In terms of Age, the largest age group of respondents within this research is 20
- below 25 (40.3%), which is followed by 25 - below 30 age group (24.3%). This
demographic profile is quite reasonable as consumer group from 20-30 typically
emerges as the major purchasing power for online food delivery, according to
Intage’s 2019 report on Vietnamese Youth Lifestyle.
54
In terms of Monthly income, most respondents have the income from

10,000,000 VND to below 15,000,000 VND on a monthly basis, representing for
47.9% of total respondents. Groups with monthly income of 15,000,000 - below
20,000,000 VND and 5,000,000 VND - below 10,000,000 VND turn out the second
and third largest groups. The figures indicate that respondents have decent or mid-
high living income, which may allow for more spending power and premiumization
when using FDAs.
In terms of Educational background, respondents already having

University/College degree account for the largest proportion, standing at more than
85%. The other groups include respondents holding degree for Master, high school
education and Doctor of Philosophy (PhD) or Post Doc, which represent 7.9%, 5%
and 1.5% respectively. The majority of respondents with high education background
could be seen as an advantage for the research because this may probably ensure a
certain higher level of common understanding and knowledge as to mobile
application and also enhance the likelihood of submitting well-thought responses
with higher accuracy.
4.1.2.2. Descriptive statistics analysis by measurement scales

In-depth descriptive statistics of all 41 measurement items were exhibited in
Appendix 16. Adopting the five-point Likert scale, users’ evaluation on various
aspects of FDAs and usage experiences can be quantified. As a rule of thumb, if an
observed scale has its mean value of below 3, this certain factor should be
immediately improved so as to enhance user’s experience. On the other hand, as
particular observed scales have mean values being equal or more than 4, there are
high chances that users are satisfied with this, hence these factors should be
maintained or changed for the better. A quick glance at the mean values reveals that
users’ attitude generally gravitates towards the positive side, with all mean values
being larger than 3.5.
As regards factors proposed to directly/indirectly influence CI of Millennials

FDA users, their mean values lie within range from 3.5 to 4.5, in which the lowest
value is 3.59 and the highest value is 4.45. The measurement item with highest mean
55
value is FC4 (“I feel comfortable using FDAs”), indicating that FDA users find
almost no difficulty in using and interacting with the applications with all available
organizational and technical knowledge and support. Given that, it can be concluded
that there is a high extent of agreement amongst the respondents over measurement
items included in the questionnaire. Plus, these items also have certain impacts in
formulating behavioral continuance intention to use FDA of the Millennials in HCM
city.
As regards the variable of continuance intention of Millennials FDA users,
mean values for its measurement items are all larger than 4.0 – particularly, 4.06 for
CI1, 4.10 for CI2, 4.13 for CI3, 4.11 for CI4. Such high indexes indicate strong
intention of FDA users to reuse, continue using or/and use FDA(s) in the longer term.
4.2. Cronbach’s Alpha coefficient analysis
The first-run result of Cronbach’s Alpha coefficient analysis was exhibited in
Appendix 17, in which all the constructs and measurement items well met two
reliability requirements and were validated to indicate strong internal consistency.
Specifically, Cronbach’s Alpha coefficient analysis findings point out that the
Corrected Item — Total Correlation of all measurement items are larger than 0.30,
in which almost all items have this index of over 0.60 and even two lowest Corrected
Item — Total Correlation indexes are as high as 0.594 and 0.598.
In regards to the Cronbach’s Alpha coefficients, these indexes of 11 variables

are shown to be all above 0.70, with the highest index being up to 0.873 (CI). There
are 10 out of 11 variables that have Cronbach’s Alpha coefficient within the range of
0.80 – 0.95, particularly, CI, SI, COF, SA, PE, HB, PO, HM, FC, EE, in the
descending order. These figures denote a very good level of internal consistency of
variables for the research model (DeVellis, 1991). Plus, the lowest Cronbach’s Alpha
coefficient obtained is also as high as 0.794 (TFF), indicating a highly respective
extent of reliability (DeVellis, 1991). What should also be noted is that the index for
“Cronbach's Alpha if Item Deleted” of all measurement items are all lower than the
overall current Cronbach's Alpha, thereby, it can be conclude that no measurement
item should be deleted.
56
Since all indexes of Cronbach’s Alpha coefficient analysis revealed a highly

qualified extent of reliability, all the measurement scales and measurement items
proposed were all retained and proceeded into the following analytical techniques.
4.3. Explanatory Factor Analysis (EFA)

Compared to Linear Regression, there exist differences in analyzing procedure
and evaluation criteria when conducting EFA for model adopting CFA and SEM.
Table 4.1: EFA analyzing procedure and evaluation criteria for CFA/SEM-
based model
Analyzing procedure Evaluation criteria
(1) Extraction method: Principal Axis (1) Kaiser – Meyer – Olkin (KMO)
Factoring coefficient: 0.5 ≤ KMO ≤1.0
(2) Rotation method: Promax (instead of (2) Bartlett's test of sphericity: Sig <
Varimax) 0.05
(3) Suppress small coefficients: (3) Total Variance Explained: ≥ 50%
Absolute value below 0.5 (4) Factor loading: ≥ 0.5
Source: Hu and Bentler, 1999; Hair et al., 2010
4.3.1. Explanatory Factor Analysis (EFA) for 10 variables affecting Continuance

Intention (CI)
Upon the first-run of EFA, the result turns out to be satisfactory, with all
above-mentioned criteria met (shown in Table 4.2 below and Appendix 18.1).
Particularly, EFA reveals that KMO coefficient is 0.882, lying within the range of 0.5
and 1.0. Plus, Sig value of Bartlett's test of sphericity is 0.000, far smaller than 0.05,
proving that observed variables are correlated with each other as a whole. As for Total
Variance Explained, its cumulative percentage of Rotation Sums of Squared
Loadings also stands at 56.583%, above the minimum acceptable index of 50%,
which indicates that 56.583% of total variance after Promax rotation is accounted for
by 10 factors. In terms of factor loading, all 37 measurement items of these factors
have factor loading value above 0.5, with the highest being 8.36 (SI3) and the lowest
being as high as 0.641 (SA2).
57
4.3.2. Explanatory Factor Analysis (EFA) for Continuance Intention (CI)

In regards to EFA for CI, the obtained result also satisfies all the given criteria
(see Appendix 18.2). Table 4.2 clearly exhibits that KMO coefficient is 0.835 (0.5 ≤
KMO ≤1.0), suggesting the factor analysis is adequate, whilst Sig value of Bartlett's
test of sphericity is also 0.000 for qualified correlation among observed variables. In
terms of Total Variance Explained, its cumulative percentage of Rotation Sums of
Squared Loadings reaches 63.340, which indicates that 63.340% of total variance
after Promax rotation is accounted for by 01 factor (CI). Plus, factor loading of 4
observed variables for CI is all higher than 0.7, far above the minimum acceptable
level of 0.5.
Table 4.2: KMO, Bartlett's Test and Total Variance Explained

Value Value
Indicator
(for variables without CI) (CI)
Kaiser-Meyer-Olkin Measure of Sampling Adequacy .882 .835
Approx. Chi-Square 7193.219 885.562
Bartlett's Test of
df 666 6
Sphericity
Sig. .000 .000
Rotation Sums of Cumulative
Total Variance Explained 56.583 63.340
Squared Loadings %
Source: Data from IBM SPSS Statistics 20.0, 2020

From (4.3.1.) and (4.3.2.), it can be concluded that the underlying relationships
between observed variables are adequate, with all items contributing to measuring
the construct; hence, no observed variable is omitted or changed.
4.4. Confirmatory Factor Analysis (CFA)

Once the data had been processed through descriptive statistics, Cronbach’s
Alpha coefficient and Explanatory Factor Analysis (EFA), the research model
remained unchanged, with all 11 variables, 41 measurement items and 18 hypotheses
being retained. At these steps, the research model together with its hypotheses were
validated through CFA and SEM with the aid of IBM SPSS AMOS 24.0.
As regards CFA, the author performed the following steps:

58
(i) Creating a path diagram of CFA representing the factors structure via:
1. Manually drawing the path diagram with the aid of some embedded plugins;
2. Adding figure captions via the use of Macro to show key indexes on
diagram;
(ii) Calculating Estimates and checking some criteria of Model Fit, Reliability,
Convergent Validity and Discriminant Validity to validate the adequacy of CFA-
based model. Indicators of Model Fit are available in AMOS output, but other
indicators of Reliability, Convergent validity, Discriminant validity have to be
obtained through an extended plugin of “Validity and Reliability Test” developed by
Gaskin and Lim (2016).
4.4.1. CFA path diagram

Appendix 19.1 exhibits the path diagram for CFA-based model, which
represents structure of 11 variables, including: performance expectancy (PE), effort
expectancy (EE), social influence (SI), facilitating condition (FC), hedonic
motivation (HM), habit (HB), price-saving orientation (PO), perceived task-
technology fit (TTF), confirmation (COF), satisfaction (SA) and continuance
intention (CI).
4.4.2. Model-fit indices of measurement model

Table 4.3: Model-fit indices of measurement model (CFA)
Criterion Obtained value
CMIN/df ≤ 2: Good; CMIN/df ≤ 5: Acceptable CMIN/df = 1.053
GFI ≥ 0.8: Acceptable; GFI ≥ 0.9: Good; GFI ≥ 0.95: Very good GFI = 0.927
CFI ≥ 0.9: Good; CFI ≥ 0.95: Very good CFI = 0.995
TLI ≥ 0.9 TLI = 0.995
RMSEA≤0.08: Good; RMSEA≤0.03: Very good RMSEA = 0.011
PCLOSE ≥ 0.01: Acceptable; PCLOSE ≥ 0.05: Good PCLOSE = 1.000
Source: Data from IBM SPSS AMOS 24.0, 2020

59
It is apparent from the table 4.3 and Appendix 19.2 that all the model-fit
indices of measurement model (CMIN/df = 1.053, GFI = 0.927, CFI = 0.995, TLI =
0.995, RMSEA = 0.011 and PCLOSE = 1.000) respectively exceed the common
acceptance levels, particularly three indices (CMIN/df, GFI, PCLOSE) reach good
level and the other two (CFI, RMSEA) are validated to be very good. These figures
demonstrate a respectable fitness of the measurement model.
4.4.3. Validity and reliability test
As specified in table 3.2, reliability, convergent validity, discriminant validity
of the model would be tested through estimates of Standardized Loading Estimates
and the indexes of CR, AVE, MSV, SQRTAVE and Inter-Construct Correlations.
As for Standardized Loading Estimates, factor loading values extracted by 41

scale items exceed the acceptance level of 0.5 (shown in Appendix 19.3). Among
them, 36 items having factor loading of above 0.7, reaching the ideal value for
reliability test. These figures indicate all measurement items in the model are
statistically significant.
As for other indices of reliability, convergent validity (demonstrated in

Appendix 19.4), CR and AVE indexes of all variables are respectively higher than
0.7 and 0.5, with the highest CR value being around 0.874 (CI) and the highest AVE
being 0.634 (CI). Hence, the model is validated to have no concerns over reliability
and convergent validity.
Furthermore, the square roots of the AVE (SQRTAVE) values of all constructs
are greater than Inter-Construct Correlations. Plus, MSV value of a certain construct
is also lower than its respective value of AVE. Hence, discriminant validity is
established.
4.5. Structural Equation Modeling (SEM)
Since model-fit, reliability, convergent validity, discriminant validity indices
had been all validated to be adequate in the former step, SEM was then carried out to
test hypotheses about relationships amongst observed and latent variables.
Path diagram for SEM was also built out on the basis of path diagram for CFA
and was demonstrated in Appendix 20.1. Additionally, all the indices for model-fit,
60
reliability and validity indices were also statistically proved to be adequate (see
Appendix 20.2) and updated accordingly in the diagram. In terms of hypotheses
testing, there presents three key output table worth nothing, which include:
(1) Regression Weights (shown in Appendix 20.3)
As all sig values (or, p-values) are smaller than 0.05, all measurement items
are statistically significant. The magnitude of the estimate reveals the impact of one
factor on another factor within the hypothesized path. Since all the estimates are
larger than 0, it can be concluded that the given relationships between variables are
positive.
In particularly, continuance intention (CI) of using FDAs amongst Millennials

in HCM City is significantly predicted by the role of performance expectancy (PE)
(γ = 0.282, p < 0.001), effort expectancy (EE) (γ = 0.158, p = 0.023), social influence
(SI) (γ = 0.139, p=0.003), facilitating condition (FC) (γ = 0.122, p= 0.043), hedonic
motivation (HM) (γ = 0.126, p= 0.040), price-saving orientation (PO) (γ = 0.159, p=
0.007), habit (HB) (γ = 0.225, p < 0.001), and satisfaction (SA) (γ = 0.228, p < 0.001)
directly. Hence, hypotheses H1, H3, H6, H8, H10, H13, H12, H18 are confirmed.
Additionally, the statistical results illustrate that satisfaction (SA) is

significantly formulated by performance expectancy (PE) (γ = 0.211, p =0.002),
confirmation (COF) (γ = 0.268, p<0.001), effort expectancy (EE) (γ = 0.199, p =
0.023), social influence (SI) (γ = 0.130, p = 0.028), facilitating condition (FC) (γ =
0.217, P = 0.006), hedonic motivation (HM) (γ = 0.126, p= 0.040), price-saving
orientation (PO) (γ = 0.159, p= 0.007) respectively. Thus, hypotheses H2, H16, H5,
H7, H9, H11 and H14 are validated.
In the meanwhile, perceived task-technology fit (TTF) (γ = 0.221, p < 0.001),

confirmation (COF) (γ = 0.150, p=0.004) and effort expectancy (EE) (γ = 0.605, p <
0.001) are verified to have statistically significant influence on performance
expectancy (PE), which support H4, H7 and H5.
(2) Standardized Regression Weights

61
Data shown in Appendix 20.4 and Table 4.4 illustrate the standardized
estimates, where the underlying data have been standardized so that the variances of
all variables are equal to 1 (Lewis-Beck et al., 2003). A quick glance over the table
suggests the order of impact of other variables on the dependent variables, including
CI, SA and PE. As for CI, among eight factors, PE (β=0.270) is validated to be most
impactful, followed by SA (β=0.234) and HB (β=0.197). Besides, COF proves to
have the greatest influence in shaping SA with β=0.238, with the second and the third
being PE (β=0.197) and EE (β=0.183). Regarding PE, this variable is significantly
influenced by EE (β=0.595), TFF(β=0.233) and COF (β=0.42) respectively.
Table 4.4: Standardized Regression Weights
Hypothesized path Estimate (β) Hypothesized path Estimate (β)
PE <--- TTF .233 CI <--- PE .270
PE <--- COF .142 CI <--- EE .149
PE <--- EE .595 CI <--- SI .135
SA <--- PE .197 CI <--- FC .107
SA <--- COF .238 CI <--- HM .104
SA <--- EE .183 CI <--- PO .108
SA <--- SI .123 CI <--- HB .197
SA <--- FC .186 CI <--- SA .234
SA <--- HM .120
SA <--- PO .158
(3) Squared Multiple Correlations
Table 4.5 exhibits the squared coefficient of multiple correlation, or R2, that
indicates how well a given variable can be predicted using a set of other variables.
62
Table 4.5: Squared Multiple Correlations

Estimate Estimate Estimate
PE .448 SA .550 CI .738
Specifically, for CI, its R2=0.738 suggested that 73.8% of the total variation in
CI can be explained by the model, particularly by 08 factors (PE, EE, SI, FC, HM,
PO, HB, SA). Furthermore, with R2 being equal to 0.550, the independent variables
(namely, PE, COF, EE, SI, FC, HM, PO) have statistically significant explanations
for 55% of the variation of SA. Likewise, when it comes to PE, three variables of
TTF, COF, EE can predict around 44.8% of variation within PE, whose R2 stands at
0.448.
4.6. Independent Sample t-Test & One-Way ANOVA on demographic variables

4.6.1. Gender
The impact of gender on Millennials’ continuance intention to use FDA was
analyzed through Independent Samples t-Test, whose result was exhibited in
Appendix 21.1. The finding reveals that the significance level (sig) of Levene's Test
for equality of variance is 0.158, far exceeding the threshold of 0.05; hence, there is
no statistically significant differences in variance between the two genders.
Furthermore, sig of t-Test at “Equal variances assumed” stands at 0.776, which is
higher 0.05, so there is also no statistically significant difference in continuance usage
intention between two genders.
4.6.2. Age
Analytical result of One-Way ANOVA on the demographic variable of age
was demonstrated in Appendix 21.2. All valid respondents are classified into four age
groups, including (1) 20 – below 25, (2) 25 – below 30, (3) 30 - below 35 and (4) 35
– 40 (40 included). As sig of Levene's Test is around 0.439, above the threshold of
0.05, it can be concluded that there is no difference in variances between these four
age groups. Then, the author conducted F-test in ANOVA table. Regarding F-Test,
sig value reaches 0.146, which is much higher than 0.05, thus, the hypothesis that
63
differences in age result in differences in Millennials’ continuance intention to use

FDA is rejected.
4.6.3. Monthly income

Appendix 21.3 represents the result of one-way ANOVA test on monthly
income. Specifically, within this study, the author divided this variable into five
groups, which consist of (1) Below 5,000,000 VND, (2) 5,000,000 VND - below
10,000,000 VND, (3) 10,000,000 VND - below 15,000,000, (4) VND 15,000,000 -
below 20,000,000 VND, and (5) 20,000,000 VND and above. For monthly income,
its sig value of Levene's Test is measured at 0.715, above 0.05, so the author rejected
the hypothesis that there is difference in variance between five groups of income. As
regards sig value for F-test, the figure stands at 0.217, being higher than 0.05,
indicating that the impact of monthly income on Millennials FDA users’ continuance
intention is not statistically significant.
4.6.4. Educational background

Similar to Age and Monthly income, the influence of Educational background
on formulating Millennials FDA users’ continuance intention is tested through one-
way ANOVA, with the result illustrated at Appendix 21.4. This demographic variable
is divided into four groups, namely (1) High school, (2) University/College, (3)
Master and (4) Doctor of Philosophy (PhD) or Post Doc. Because sig value of
Levene's Test on Education background is 0.966, exceeding the threshold of 0.05,
there is no statistically significant differences in variance amongst the four given
groups. Furthermore, sig value for F-Test is measured at 0.246, which is far greater
0.05; hence, it can be referred that differences in educational background do not lead
to significant differences in the behavioral continuance intention to use FDA amongst
the Millennials in HCMC.
4.7. Research result discussion

4.7.1. Research result
Statistical findings have confirmed all the proposed hypotheses and validated
that eight variables - namely, PE, EE, SI, FC, HM, HB, PO, SA - were statistically
significant antecedents of Millennial users’ continuance usage intention (CI), wherein
64
PE has the most substantial influence and other variables impact CI in the following
order: PE > SA > HB > EE > SI > PO > FC > HM.
Figure 4.1: Result of research model
4.7.2. Research result discussion

With the standardized coefficient β=0.270 and β=0.197 respectively, PE
stands as the strongest determinant of CI and the second-strongest determinant of SA,
which manifests the importance of the cognitive and functional benefits of FDAs
from the HCM city Millennials customers’ perspective. There present several
appealing features provided by FDAs that allow for more mobility and flexibility in
ordering food online compared to the traditional ways such as visiting restaurants in-
person or telephoning them. As a result, FDA users, especially the tech-savvy
Millennials, are enabled to save a great deal of time and effort. Such results pertaining
to the significant influence of PE parallel those reached by Mun et al. (2017), Yeo
et al. (2017), Okumus et al. (2018), Roh and Park (2019), Palau-Saumell at al.
65
(2019), Suk Won Lee at al. (2019), Alalwan (2020) on FDAs and other kinds of food
apps such as diet apps or apps for restaurants.

EE is also validated to influence both CI and SA to a certain extent. This
finding is consistent with the survey results conducted by Q&Me (2020), which
reveals that ease of ordering food though FDAs is greater compared to dedicated
restaurant apps and ordering method via social networks. And this explains why ease
of use of FDAs stands among the top reasons shaping respondents’ satisfaction.
Whereas there exist many studies that undermine the role of EE in satisfaction,
intention to use/continuously use IS, the predictability power of this variable has been
highlighted in several other papers, including Kim and Malhotra (2005), Venkatesh
et al. (2012), Kang (2014), Fang and Fang (2016), Ray et al. (2019), Marinković et
al. (2020).

Similarly, statistical results prove the significant influence of SI on CI and SA.
As a connected generation (Taylor and Keeter, 2010), Millennial consumers, as well
as their purchasing intentions, are typically shaped by others’ opinions, especially
those deemed as important or influential to them (Kantar, 2018). Such statistically
significant relationships of SI on CI and SA have also been confirmed by Lee at al.
(2019), Palau-Saumell at al. (2019), Roh and Park (2019), Zhao and Bacao (2020)
on food apps, or Zhou and Li (2014), Lai and Shi (2015), etc. on other mobile
technologies and IS.

FC is also verified to be the determinant of CI and SA; whilst the impact of
FC on CI is not as strong as other factors, FC emerges as the third most influential
variable forming SA. Chan et al. (2010), Maillet et al. (2015) and Alalwan (2020)
also confirmed that the availability of technical and human support performs a critical
role for consumers to be pleased about their experience of adopting FDAs. Compared
to SA, FC leaves fewer impact on CI, which could be attributed to the fact that FC
are more related to the users’ immediate experience (i.e. actual usage, satisfaction,
66
etc) (Chan et al., 2010; Maillet et al., 2015) rather than long-term continuance
intention (Alalwan, 2020). This is especially true for actual adopters who are digital-
savvy and obtain adequate experience with technology (Mathieson, 1991; Venkatesh
et al., 2003, 2012; Okumus et al., 2018) – who is also the target respondents of this
study.

Although HM emerges as the least influential factor, this factor still performs
an indispensable part in formulating FDA users’ sense of satisfaction as well as their
decision to continue using or reject these food platforms. From psychological
perspectives, those, who derive a sense of pleasure and enjoyment from the usage
experience of FDAs, are more likely to keep using these applications in future (Davis
et al., 1992; Van der Heijden, 2004). According to Alalwan (2020), significant
influence of HM is partly attributable to the novelty and innovativeness of FDAs,
which may expedite the users’ sense of pleasure and enjoyment. Considering the
context of Vietnam, especially in HCM city, where FDAs have been widely adopted,
particularly among convenience-driven Millennials, the role of HM may not be as
strong as empirically investigated. This finding reaches consensus with Lee at al.
(2019).
4.7.2.6. Habit (HB)

HB was statistically verified to be the third strongest predictor of CI. Such
significant role of HB on continuous use intention has been empirically validated by
Palau-Saumell et al. (2019) and Lee at al. (2019) on FDAs’ continuance usage
intention, along with Morosan and DeFranco (2016), Amoroso and Lim (2017),
Rana et al. (2017), etc. on mobile applications and e-commerce. Within today’s
ever-evolving mobile landscape, individuals are increasingly attached to their
smartphones and develop a habitual behavior towards using the associated mobile
apps. And once they have formulated a habitual pattern towards using a new system,
there is every likelihood that they retain their motivation to adopt the system in future
(Amoroso and Lim, 2017; Sun and Chi, 2018). Within this research, there presents
a clear pattern that respondents, who claimed to order food many times through FDAs
67
(“more than 20 times per month’”) as a natural habit, are ones having greater
measurement items underlying CI.

As an extended variable replacing Price Value construct of the original
UTAUT2, PO is confirmed to exert statistically significant impacts on both SA and
CI. Compared to traditional approaches of having meals, modern consumers do prefer
the availability of information on products and prices on FDAs so as to facilitate
comparison. Besides, several price-conscious users actively seek for discounts and
special offers to such an extent that they feel FDAs more of a saving option of
having meals. The statistically significant predictive power of PO has also been
confirmed by several papers, including ones by Palau-Saumell et al. (2019) on
restaurant apps, Dazmin (2019) on food delivery services and Akroush and Al-
Debei (2015) on online shopping in general.
4.7.2.8. Perceived Task-Technology Fit (TTF)

As a pivotal construct extracted from the task-technology fit model, TTF is
verified to perform a significant role in formulating users’ PE in technology
adoption. With FDAs’ remarkable features, typically real-time tracking of food
delivery, their functions meet users’ requirements of accessing meals and/or food
supplies, making users to perceive FDAs to be useful and compatible. This result
replenishes findings from Kurniawati (2019), Roh and Park (2019), Zhao and
Bacao (2020) that TTF is another antecedent of PE towards continuance usage of
FDAs.
4.7.2.9. Confirmation (COF)

Not only TTF and EE but also COF is validated to be a notable predictor of
PE, despite having a quite moderate impact with estimate parameter of 0.142. In
synchrony, this variable also significantly determines SA, being validated to have
strongest power of predictivity for SA. As COF reflects users’ expectations
formulated by their previous experience of IS use, the actual realization of
expected benefits will definitely formulate a sense of satisfaction towards the
68
continuance usage intention. Such statistical findings pertaining to the significant

impact of COF on PE and SA parallel those proposed by Lee and Kwon (2011), Yuan
et al. (2016), Tam et al. (2018), Alshurideh et al. (2020), Zhao and Bacao, (2020).
Under the influence of seven factors (namely, PE, COF, EE, SI, FC, HM, PO),
SA concurrently emerges as the second-strongest predictor shaping continuance
intention to use FDAs among Millennials in HCM city. Understandably, if consumers
felt content with the outcomes of their prior behavior and experience, they intend to
keep repeating such behavior – and continuance usage of FDAs is not an exception.
Empirically, the role of SA as the top motivation for FDAs usage continuance has
been supported by a plethora of previous papers in various fields of IS and mobile
technologies, including Abdallah Alalwan (2020) and Zhao and Bacao (2020) on
FDAs.
In this chapter, statistical analyses were respectively performed. Firstly, the
author leveraged descriptive statistics on demographic factors and measurement
items to quantitatively summarize features from garnered data. Cronbach’s Alpha
coefficient analysis was then carried out, which validated the reliability of all the
measurement items to a highly qualified extent. EFA also delivered a satisfactory
result, in which all underlying relationships between observed variables were
adequate and no construct/item got omitted or changed. Afterwards, CFA and SEM
were performed with the aid of IBM SPSS AMOS platform. CFA-based and SEM-
based path diagram were built out, with all indices of model fit, reliability and validity
being statistically significant. Notably, the results of the path estimate coefficients’
analysis supported all the hypotheses, with performance expectancy (PE) validated
to be the most influential factor shaping continuance intention to use FDA amongst
the Millennials in HCM city, which was followed by SA, HB, EE, SI, PO, FC, HM
respectively.
69
CHAPTER 5: RESEARCH CONCLUSION AND RECOMMENDATIONS

5.1 Research conclusion
Analytical findings confirmed the validity and reliability of all measurement
scales as well as validated all proposed hypotheses. 08 proposed factors (namely, PE,
EE, SI, FC, HM, HB, PO, SA) could explain 73.8% of the total variation in CI, which
indicated the respectable predictive power and applicability of research model on the
field of technology continuance usage. Other two variables of TTF and COF were
also statistically proved to indirectly formulate CI via their relationships with PE
and/or SA. On this ground, the author proposed some recommendations for key
market participants, including FDA providers, F&B merchants, traditional
restaurateurs and policy makers. Accordingly, this study contributes with some
theoretical and practical implications:
Regarding theoretical implications, the research enriched the literature of

technology continuance usage, especially on the topic related to FDAs, which was
still at the nascent stage. Integrating UTAUT2, ECM and TFF model for first time,
the study proposed a comprehensive model that was validated to highly explain
technological and mental antecedents of FDA users’ continuance intention.
Regarding practical implications, this study presented practical understanding

as to key determinants shaping users’ efficiency expectancy, sense of satisfaction,
and especially their FDA usage continuance intention, from which improvement
rooms for key parties involved in food delivery market were respectively proposed.
5.2. Research recommendations

5.2.1. Recommendations for food delivery app providers
Considerable emphasis should be put into the role of PE, which was
statistically confirmed to be the most influential antecedent formulating Millennial
FDA users’ continuance intention, as well as the second-strongest determinant of
users’ satisfaction. Acknowledging such significance, the author proposed the
following recommendations to enhance users’ perception of the usefulness and
efficiency of adopting.
70
Firstly, there should be more focus on speediness and accuracy in informing

delivery time on food platforms. Large-scale surveys conducted by GCOMM (2019),
Havas Riverorchid (2019) and Q&Me (2020) have consistently confirmed the role of
delivery as the most decisive factor shaping FDA users’ satisfaction and intention to
use frequently. Furthermore, in the questionnaire survey by the author, some
respondents also suggested improving delivery speed for more satisfied usage
experience. Indeed, issues related to delivery time, such as slow delivery speed or
unreasonable differences between expected and actual delivery time, are likely to
affect user’s perception of FDA usefulness and result in users’ dissatisfaction
accordingly (Q&Me, 2020).
Secondly, FDA providers should increasingly invest efforts into attracting

more and more F&B partners, which aims to enhance the diversity of food portfolios.
Another noteworthy point is that, besides the option of partnering with external
restaurateurs, FDA providers can themselves develop their own cooking facilities so
that they have full control over the whole process from preparing food to deliver
meals to users’ doorsteps.
According to Statista’s report (2020), local cuisine, fast food, tea-based drinks
(especially bubble-tea) and snacks are the top 4 food types preferred by both male
and female FDA users. Among them, male users order local cuisine the most, whereas
it’s fast food that emerges as the most favorite type chosen by female. Given that,
FDA providers should establish adequate growth strategy and practices to win more
partners and ultimately more users. Additionally, over the recent years, especially
after the burst of COVID-19, the Vietnamese take keeners interest in adopting healthy
eating habits and lifestyle (Nielsen, 2020). Accordingly, they should expect a wider
range for “clean and green” food choices as well as more accompanying nutritional
information. Hence, FDA providers shall promptly work with their partners and
address these needs.
Thirdly, besides huge base of F&B partners, food quality provided by these
partners should be taken into close consideration. Compared to eating out, ordering
food online does not allow consumers to see the product firsthand before making
71
purchasing decision; thus, authentic information and image for specific meal portion
are “must-haves”. Poor quality of food not only hurts food stall’s reputation but also
indirectly affect users’ perception over FDAs in a negative manner (Q&Me, 2020).
Fourthly, since convenience is a major advantage of FDAs in comparison with

traditional ways of having meals, FDA providers should consider launching
marketing campaigns to convince and continuously remind customers that FDAs
require minimal time and effort. Such a marketing message is expected to matter the
most for convenience-driven and time-poor audience, particularly the Millennials.
There remain other proposals given to enhance FDA users’ PE, which will be
discussed further in the sections concerning EE, TTF, COF.

Considering that EE is statistically confirmed to perform certain role in
shaping FDA users’ PE, SA and CI, FDA providers should also put more thoughts
into this factor. A couple of suggestions were put forth as follows:
Firstly, intuitive and user-friendly technical design should be applied for

FDAs. FDAs’ interface, functional features, or in-app navigation should be designed
in a way by which users can effortlessly and instantly have what they look for.
Although the Millennials are characterized by being tech-savvy, a brief guide on FDA
usage is always indispensable. Plus, further app update or continued development of
other functionalities are welcomed, yet should not require much effort and adaptation
from users.
Secondly, online payment options, typically via e-wallet and internet banking,
should be integrated into the system. Millennials embrace the ease of using mobile
payments, especially in online shopping, including ordering food (Bermeo-Giraldo et
al., 2019; Facebook and Bain & Company, 2020). So far, while Now, Grab Food and
most recently Baemin have incorporated these cashless payment methods into their
platforms, other players, especially GoFood, still solely depend on cash-on-delivery
approach. Prompter action should be taken for the given FDA providers as well as
any later market entrants.
72

Since statistical results also prove the significant influence of SI on Millennial
FDA users’ sense of satisfaction and their continuance intention, FDA providers
should acknowledge its role and strengthen this factor to attract more users. In regards
to this, there were two recommendations proposed by the author:
Firstly, FDA providers may consider launching referral and affiliate programs.
FDA users should be incentivized to refer the platforms to their friends, colleagues
or relatives in exchange of one-time referral payment or bonus; meanwhile, FDA
providers can also run some affiliate programs, partnering with relevant affiliate sites
in the mechanism of commissions. In fact, several mobile applications, typically e-
wallet MOMO, have leveraged these tactics quite successfully to win users.
Secondly, key opinion leaders, better known as KOLs, could be invited to join
the marketing campaigns. Within Vietnamese market, there have been some FDA
providers adopting this approach, including: GrabFood - famous singer My Tam and
list of micro-influencers; GoFood - first-class singer Son Tung M-TP and recently
Hari Won; Baemin - well-known comedian Tran Thanh. Although the idea is not
novel, such a strategy of influencer marketing is especially influential to generation
Y (Morning Consult, 2019).

In view of significant impact of FC, it is of great necessity to pay attention to
the technical and human support required to facilitate users’ access and use of FDAs.
Firstly, for human support, FDA providers should have their customer service
system available 24/7, ensuring that any request or concern is handled without delay.
Secondly, for technical aspects, FDAs should be designed to be more

compatible with and comparable to any other mobile applications that users already
use. Besides, organizational attempts to develop innovative supporting features, such
as customized help, interactive communication channels, FAQs (Frequently Asked
Questions), should be made, aiming to help users obtain any assistance or information
at any time they need.
73

Although statistical findings of this research reveal that HM is less influential
to Millennial FDA users’ satisfaction and continuance compared to other factors, an
adequate effort should be invested so as to promote its positive role. In terms of the
hedonic aspect of FDAs, the author put forward the two following proposals:
Firstly, since users’ sense of pleasure when using FDAs partly comes from the
novelty and innovativeness of these platforms (Alalwan, 2020), developers may
consider continuously innovating these apps’ design along with developing
entertaining mini-games, so as to constantly deliver enjoyable users’ experience.
Secondly, focus of promotional content within a long-run marketing campaign

may be directed towards various facets of hedonic utilities associated with using
FDAs. Specifically, marketers can craft some content pillars highlighting the
interesting aspects of FDA usage, conveying attractive messages that using FDA
continuously is a natural part of the modern lifestyle. To bring such campaigns to life,
some selective social media platforms, which are popular among the Millennials,
such as Facebook, Instagram, YouTube, are recommended to be leveraged.

Price-saving orientation, typically referred to as the availability of discounts
and capacity for price comparison among various food choices, is statistically
validated within this study as well as several other papers to be a strong determinant
of users’ satisfaction and usage continuance intention. On such basis, the author
recommended:
Firstly, further financial incentives (i.e. deals, quantity discounts, price

discounts, accumulative points and rewards like vouchers), combined with loyalty
programs should be promoted for FDA active users. This research’s survey result
reveals that discounts capture substantial amount of attention from respondents and
strongly impact on their FDA choice. Besides, special offer assumes a remarkable
role to loyal users; when properly adopted, this tactic can expedite users’ loyalty,
continuance usage intention and even their willingness to give recommendations to
others in exchange of a worthy reward
74
Secondly, FDA providers and their F&B partners should collaborate on how
to establish the most optimal pricing scheme for food items. There exist certain
comments mentioning that sometimes, meal ordered via FDAs has smaller portions
compared to food ordered conventionally, and that price of food items provided by
certain restaurants via delivery apps is set higher than that specified in-store. To
maintain repeated purchases of FDA users, these concerns should be quickly and
profoundly addressed.
Charging fee for using FDAs is also strongly discouraged as additional

expenses incurred by app users are likely to result in dissatisfaction and discontinued
usage. This brings no incremental benefits to FDA providers in the long run.
5.2.1.7. Habit (HB)

Emerging as the third strongest predictor of Millennial FDA users’
continuance usage intention, HB should receive substantial attention from FDA
providers. With an aim of developing habitual usage pattern of FDA users, some
suggestions were put forth:
Firstly, FDAs should offer diverse services to encourage users to reinforce the
habit of using FDAs on a regular basis. Additional activities, typically loyalty
programs, discounts, special offers and tastings, may help users to become more
receptive and ultimately lead to FDA usage being a natural part of their daily routine.
Secondly, some technical features can be embedded into FDA systems so as

to gradually shape users’ usage patterns. Based on recent reports by Intage Vietnam
(2020) and Q&Me (2020), lunch and teatime are two occasions when food delivery
is most utilized, particularly among the age group of 20-30. Given that, FDA
providers may consider adding push notifications, which recommend users’ favorite
food choices for lunch, or offer discounts tailored uniquely to food order at this time.
5.2.1.8. Perceived Task-Technology Fit (TTF)

Mostly addressing the technical aspects of FDA adoption, perceived task-
technology fit was also validated to perform a significant role in formulating users’
perception of usefulness. Thus, the author presented three proposals as follows:
75
Firstly, with a view to facilitating users’ efficient access and ordering of food
items through FDAs, regular system monitoring, maintenance and update to verify
the reliability and platform quality of FDAs should be necessitated. Some common
issues of mobile apps such as slowing loading speeding and overwhelmed capacity
at peak hours (as claimed by this study’s respondents), should be timely detected and
resolved.
Secondly, since order tracking is considered one of the most recognizable and
critical features of FDAs, numerous considerations should be taken into account in
order to design a full-stack tracking system. Serving the purpose of providing users
with a visual and easy way to monitor the progress of their orders, map tracking tends
to be most compatible and also the most commonly adopted approach. Plus, other
relevant pieces of information regarding food orders, such as estimated delivery time,
location of the restaurants, distance between restaurant and customer’s location,
product quantity, cost, name of the driver as well as his/her contact number, should
be clearly informed to users. It is also of great importance to ascertain that all the
information given by such an online tracking system is accurate, reliable, and
credible; otherwise, there are high chances that FDA users lose trust in the online
tracking ability and in FDAs as a whole.
Thirdly, in addition to basic functional capacities, FDA providers can take use
of major advantages of mobile technology, such as personalization, responsiveness,
ubiquitous connectivity, and active control, to design further advanced technical
features with a view to enhancing users’ experience of ordering food.
5.2.1.9. Confirmation (COF)

Similar to TTF and EE, confirmation, or the realization of expected benefits
of FDA, is influential to users’ perception of usefulness and efficiency. This factor
also proves to be a notable antecedent of Millennial FDA users’ satisfaction.
Considering its significance, three following suggestions were put forth to enhance
the role of this factor:
Firstly, FDA providers should enhance service quality as well as leverage

technological advancements to deliver personalized experience for users. Once users
76
are taken individually and provided with several offerings tailored uniquely to their
specific preferences and tastes, their expectancy over FDAs’ performance and
usability would be fulfilled and accordingly their satisfaction would be elevated. A
couple of proven personalization practices include: (1) to allow FDA users to
customize their own dishes or submit specific mix-in requests, or (2) to provide
personalized offers based on users’ particular interest, order history, location, or on
special occasions, like their birthdays.
Secondly, FDA users should be motivated to give review and/or rating after
each order, which then should be promptly proceeded and thoroughly addressed. No
matter how simple this idea seems to be, relevant and credible reviews and ratings
are of utmost importance for FDA providers to recognize users’ concerns, from which
they adjust or revamp their services accordingly. Certain critical aspects of food
delivery, such as delivery speed, food quality, food cost and drivers’ attitude (Q&Me,
2020), should be particularly emphasized and requires considerable attention.
Thirdly, close consideration should be taken into information quality. Given

that differences from claimed information and reality can result in users’
dissatisfaction, FDA providers should make sure that all information given is of the
highest quality in terms of accuracy and timeliness. To achieve this objective, FDA
providers should work closely with their F&B partners so as to frequently check and
confirm the reliability of such information as business hours, the availability of menu
food items and price fluctuation. Any adjustments should be immediately updated in
order to minimize users’ inconvenience and complaints.

Undoubtedly, users’ sense of satisfaction derived from previous usage
experiences performs an important role in formulating their intention to reuse FDA.
Thus, monitoring and enhancing users' satisfaction by thoroughly understanding their
technological and mental expectations, as well as consistently delivering great service
quality to meet those expectations are prerequisite approaches for FDAs providers to
retain customers and sustainably ignite future development of their platforms.
77
Since satisfaction is confirmed to be shaped by the seven given factors

(namely, PE, COF, EE, SI, FC, HM, PO), strengthening the role of these factors
comprehensively and simultaneously is the best way to fostering satisfaction.
5.2.2. Recommendations for food & beverage partners on food delivery platforms
To establish reputation and ignite their profitability with FDAs, F&B
merchants should also orient efforts to enhance the role of the above antecedents.
Regarding their relationship with FDA users, they should particularly lay
strong emphasis on monitoring food and information quality, together with creating
an adequate pricing mechanism. Within an ever-evolving competitive landscape
within FDAs, each merchant should develop a wider food portfolio to satisfy diverse
needs, as well as continuously strengthen their competitive advantages, whether in
quality or price, to attract more food orders. To capture attention from FDA users, it
is also crucial to invest in taking appealing images and offering essential information
for each food item.
As for their relationship with FDA providers, two parties should always keep
each other informed and work closely to form the most optimal solutions for both. In
addition to following guideline provided by FDA providers, F&B merchants should
leverage assistance and consultancy from these platforms – in terms of market
research, packaging, technology-based management, to name a few - to resolve their
business challenges and optimize their performance.
5.2.3. Recommendations for traditional restaurants

Since food delivery segment have gained a firm foothold, and more consumers
- especially convenience-seeking Millennials - take up the habit of ordering food
online, in-store dining, together with foot traffic in traditional restaurants with a
physical storefront, witnesses continuous decreases. To adapt to such changing
consumer demand and business landscape, traditional players within the F&B sector
should react quickly by embracing food delivery service and integrating it into their
operating mechanism. In addition to reaching huge customer base, boosting online
visibility and obtaining more order, joining food delivery platforms enables
restaurateurs to cut down their overheads while also streamlining cumbersome
78
processes such as renting location, spending on advertising costs, and developing

shipper workforce.
Apparently, the power of food delivery is most evident during the lockdown
amid COVID-19 outbreak, when online food delivery was well-acknowledged for
enabling many food businesses to survive and flourish.
5.2.4. Recommendations for policy makers

In the light of its blossoming growth prospect, food delivery sector has created
certain remarkable impacts, which can be outlined into three key pillars as follows:
Table 5.1: Economic, social and environmental impacts of food delivery sector
Impacts of food delivery
Economic Social Environmental
Revenue source Comfort & convenience Plastic waste
Business opportunities for Unhealthy eatery habits & Food waste
supporting/complementary sedentary lifestyles Carbon footprint
industries Interpersonal relationships
Ample employment opportunities Challenges to public traffic
Low job satisfaction of drivers Lifeline amid COVID-19
New business model & concept
Source: Charlene Li et al., 2020
In the face of these impacts, policy makers perform increasingly important

roles, which are exhibited from two aspects:
5.2.4.1. Accelerating the growth of food delivery sector

Since food delivery brings out tremendous economic gains, especially a huge
source of revenue and employment opportunities for the population, together with
better serves the demand for convenience of modern consumers, governmental
agencies should introduce proper incentive programs to expedite the growth of this
sector. Favorable conditions, such as low-interest business loans, should be facilitated
for both FDA providers and F&B merchants. Synchronously, supporting
infrastructure for food delivery, including telecommunication network and
79
transportation system, should also be strengthened to address the current traffic-

related challenges as well as further contribute to the public interest.
5.2.4.2. Managing unexpected impacts of food delivery sector

To handle negative economic impacts of food delivery, policy-makers should
take measures to establish a solid framework to promote adequate working conditions
for delivery personnel in attempts to diminish attrition rates and accordingly foster
economic sustainability. From social standpoint, policy-makers may raise public
awareness of sustainability, healthy eating habits, along with the harm of sedentary
lifestyles through education. Such educative programs can be launched in cooperation
with schools, companies, social establishments while also being promoted via public
media means.
Special attention should also be paid to environmental aspects, in which

decision-makers can implement relevant policies clarifying the rights and
responsibilities of each stakeholder (i.e. FDA providers, F&B partners, FDA users)
in waste and carbon footprint. According to the Ministry of Natural Resources and
Environment (MONRE), every day around 80 tonnes of plastics is discharged in
Hanoi and Ho Chi Minh City alone, which is contributed partly by food delivery
sector. Yet, these market operators have yet to offer any real solutions on the issue.
According to Vietnam Investment Review (VIR) - a news site under the management
of Ministry of Planning and Investment, GrabFood and Loship seem to be the only
two food delivery platforms engaged in activities to relieve plastic waste burden, yet,
their efforts remain feeble and change the habits of only a fraction of consumers and
partners. Thus, stronger course of action should be taken to timely and profoundly
tackle these given environmental issues.
5.3. Limitations and proposed future research orientation

Regardless of marked theoretical and practical implications on FDA users’
continuance intention, this study inevitably contains some limitations to be noted,
which, yet, opens potential research avenues for future research.
Firstly, due to time constraint and lack of manpower, this research adopted
convenience sampling method; thus, its generalizability to represent whole target
80
population is limited and the estimates derived from these samples may be biased
(Jager et al., 2017). Thus, to address this discrepancy, future researchers should
employ random sampling techniques; besides, wider research scopes (maybe
including other key cities such as Ha Noi, Da Nang, Can Tho, considering regional
cultural differences), larger sample size, along with longer duration for conducting
research are highly recommended so as to extrapolate to the target population and
produce more representative results.
Secondly, the research focuses on the FDA users’ perspectives only, not
approaching other organizational market participants such as FDA providers or F&B
merchants. For this reason, some insightful data may not be taken into account and
hence, certain proposals by the author may not be adequate. Future studies should
invest time in gathering additional information from these groups, as well as focus on
further aspects concerning platform owners and delivery personnel engaged with
FDAs.
Thirdly, this study approached third-party FDAs as a whole, without

distinguishing the differences among various platforms such as GrabFood, Now,
GoFood, Baemin or Loship, which may be subject to certain level of inconsistency
in responses by users of different FDA platforms. Such differences should be put into
perspectives in future studies to deliver more practical managerial implications for
one specific FDA provider.
Within this chapter, the author presented brief research conclusion
summarized from analytical findings in chapter 4, together with some theoretical and
managerial implications contributed by the research. Based on statistical results, the
author respectively proposed some recommendations for key parties involved,
including FDA providers, F&B merchants, traditional restaurants and policy makers.
Proposals for FDA providers were put forward on the basis of factors affecting
continuance intention to use FDA amongst the Millennial users. Finally, limitations
of this study were acknowledged, which offered some revenues for future research
lines.
81
REFERENCES
I. Domestic references
1. B & Company, 2020, Third-party delivery service is emerging in food

delivery market.
2. Deloitte, 2020, The Vietnam Consumer Survey: An accelerating momentum.
3. Euromonitor, 2018, 100% Home Delivery/Takeaway in Vietnam,
Euromonitor International.
4. Fitch Solutions Vietnam, 2020, Vietnam Food and Drink Report Includes 5-
year forecasts to 2024.
5. Google, Temasek, Bain, 2019, e-Conomy SEA 2019.
6. Hoang Trong and Chu Nguyen Mong Ngoc, 2008, Phân tích dữ liệu nghiên
cứu với SPSS. NXB Hồng Đức.
7. IMARC group, 2020, Vietnam Online Food Delivery Market: Industry
Trends, Share, Size, Growth, Opportunity and Forecast 2020-2025
8. Intage Vietnam, 2019, What's Shaping the Youth Lifestyles.
9. Kantar Worldpanel, 2018, Vietnam: Into the minds of Millennial shoppers.
10. Kim Dang, A., Xuan Tran, B., Tat Nguyen, C., Thi Le, H., Thi Do, H., Duc
Nguyen, H., Hoang Nguyen, L., Huu Nguyen, T., Thi Mai, H., Dinh Tran,
T. and Ngo, C., 2018, Consumer preference and attitude regarding online
food products in Hanoi, Vietnam, International journal of environmental
research and public health, 15(5), p.981.
11. Nguyễn Đình Thọ, 2012, Phương pháp nghiên cứu khoa học trong kinh
doanh, NXB Lao động Xã hội.
12. Nielsen, 2018, The Nielsen Vietnam Omni-Shopper Trend Report 2018.
13. Q &Me, 2020, Corona virus impact on Vietnamese behaviors, Asia Plus Inc.
14. Q &Me, 2020, Food delivery demand increase after Covid-19, Asia Plus
Inc.
15. Q &Me, 2020, Vietnam mobile app popularity 2019, Asia Plus Inc.
16. We are Social and Hootsuite, 2020, DIGITAL 2020 – Vietnam, Datareportal
Data.
82
II. Foreign references
1. Aarts, H. and Dijksterhuis, A., 2000, Habits as knowledge structures:

Automaticity in goal-directed behavior, Journal of personality and social
psychology, 78(1), p.53.
2. ACCENTURE STRATEGY, 2017, THE FUTURE OF FOOD: New realities
for the industry, The Voice of Retail.
3. Agrebi, S. and Jallais, J., 2015, Explain the intention to use smartphones for
mobile shopping, Journal of retailing and consumer services, 22, pp.16-23.
4. Ain, N., Kaur, K. and Waheed, M., 2016, The influence of learning value on
learning management system use: An extension of UTAUT2, Information
Development, 32(5), pp.1306-1321.
5. Ajzen, I. and Fishbein, M., 2005, The influence of attitudes on behavior, The
handbook of attitudes, p. 173–221.
6. Ajzen, I. and Madden, T.J., 1986, Prediction of goal-directed behavior:
Attitudes, intentions, and perceived behavioral control, Journal of
experimental social psychology, 22(5), pp.453-474.
7. Ajzen, I., 1991, The theory of planned behavior, Organizational behavior and
human decision processes, 50(2), pp.179-211.
8. Ajzen, I., 2011, The theory of planned behaviour: Reactions and reflections,
Journal Psychology & Health, pp. 1113-1127.
9. Ajzen, I.F. and Fishbein, M., M.,1980, Understanding attitudes and
predicting social behavior, Englewood Cliffs
10. Akroush, M.N. and Al-Debei, M.M., 2015, An integrated model of factors
affecting consumer attitudes towards online shopping, Business Process
Management Journal.
11. Alagoz, S.M. and Hekimoglu, H., 2012, A study on tam: analysis of customer
attitudes in online food ordering system, Procedia-Social and Behavioral
Sciences, 62, pp.1138-1143.
12. Alalwan, A.A., 2018, Investigating the impact of social media advertising
features on customer purchase intention, International Journal of Information
Management, 42, pp.65-77.
83
13. Alalwan, A.A., 2020, Mobile food ordering apps: An empirical study of the
factors affecting customer e-satisfaction and continued intention to reuse,
International Journal of Information Management, 50, pp.28-44.
14. Alalwan, A.A., Dwivedi, Y.K. and Rana, N.P., 2017, Factors influencing
adoption of mobile banking by Jordanian bank customers: Extending
UTAUT2 with trust, International Journal of Information Management, 37(3),
pp.99-110.
15. Alalwan, A.A., Dwivedi, Y.K., Rana, N.P. and Williams, M.D., 2016,
Consumer adoption of mobile banking in Jordan, Journal of Enterprise
Information Management.
16. Alghamdi, A., Elbeltagi, I., Elsetouhi, A. and Yacine Haddoud, M., 2018,
Antecedents of continuance intention of using Internet banking in Saudi
Arabia: A new integrated model, Strategic Change, 27(3), pp.231-243.
17. Algharabat, R., Rana, N.P., Dwivedi, Y.K., Alalwan, A.A. and Qasem, Z.,
2018, The effect of telepresence, social presence and involvement on
consumer brand engagement: An empirical study of non-profit
organizations, Journal of Retailing and Consumer Services, 40, pp.139-149.
18. Aljukhadar, M., Senecal, S. and Nantel, J., 2014, Is more always better?
Investigating the task-technology fit theory in an online user context,
Information & Management, 51(4), pp.391-397.
19. Alshurideh, M., Al Kurdi, B., Salloum, S.A., Arpaci, I. and Al-Emran, M.,
2020, Predicting the actual use of m-learning systems: a comparative
approach using PLS-SEM and machine learning algorithms, Interactive
Learning Environments, pp.1-15.
20. Amin, M., Rezaei, S. and Abolghasemi, M., 2014, User satisfaction with
mobile websites: the impact of perceived usefulness (PU), perceived ease of
use (PEOU) and trust, Nankai Business Review International.
21. Amoroso, D. and Lim, R., 2017, The mediating effects of habit on
continuance intention, International Journal of Information Management,
37(6), pp.693-702.
84
22. An, L., Han, Y. and Tong, L., 2016, May, Study on the factors of online
shopping intention for fresh agricultural products based on UTAUT2,
Information Technology and Mechatronics Engineering Conference (ITOEC
2016), Atlantis Press, 2.
23. Anderson, E. W., & Sullivan, M. W. (1993), The antecedents and
consequences of customer satisfaction for firms, Marketing Science, 12(2),
125–143.
24. Anderson, J.C. and Gerbing, D.W., 1988, Structural equation modeling in
practice: A review and recommended two-step approach, Psychological
bulletin, 103(3), p.411.
25. Arenas Gaitán, J., Peral Peral, B. and Ramón Jerónimo, M., 2015, Elderly
and internet banking: An application of UTAUT2, Journal of Internet
Banking and Commerce, 20 (1), 1-23.
26. Baabdullah, A.M., Alalwan, A.A., Rana, N.P., Kizgin, H. and Patil, P., 2019,
Consumer use of mobile banking (M-Banking) in Saudi Arabia: Towards an
integrated model, International Journal of Information Management, 44,
pp.38-52.
27. Bagozzi, R.P., 2007, The legacy of the technology acceptance model and a
proposal for a paradigm shift, Journal of the association for information
systems, 8(4), p.3.
28. Bargh, J.A., 1989, Conditional automaticity: Varieties of automatic influence
in social perception and cognition, Unintended thought, 3, pp.51-69.
29. Barki, H., Titah, R. and Boffo, C., 2007, Information system use–related
activity: an expanded behavioral conceptualization of individual-level
information system use, Information Systems Research, 18(2), pp.173-192.
30. Baveja, S.S., Rastogi, S., Zook, C., Hancock, R.S. and Chu, J., 2000, The
Value of Online Customer Loyalty and how you can capture it, eStrategy
brief.
31. Bearden, W.O. and Crockett, M., 1981, Self-monitoring, norms, and attitudes
as influences on consumer complaining, Journal of Business Research, 9(3),
pp.255-266.
85
32. Bentler, P.M. and Speckart, G., 1979, Models of attitude–behavior relations.
Psychological review, 86(5), p.452.
33. Bermeo-Giraldo, M.C., Valencia-Arias, A., Duque García, B., Garcés-
Giraldo, L.F. and Luna-Ramírez, T., 2019, Factors of Use of Mobile Payment
Means in Millennials and Centennials, Semestre Económico, 22(53), pp.77-
102.
34. Bhattacherjee, A., 2001, An empirical analysis of the antecedents of
electronic commerce service continuance, Decision support systems, 32(2),
pp.201-214.
35. Bhattacherjee, A., Perols, J. and Sanford, C., 2008, Information technology
continuance: A theoretic extension and empirical test, Journal of Computer
Information Systems, 49(1), pp.17-26.
36. Bishop, P.A. and Herron, R.L., 2015, Use and misuse of the Likert item
responses and other ordinal measures, International journal of exercise
science, 8(3), p.297.
37. Boomsma, A. and Hoogland, J.J., 2001, The robustness of LISREL modeling
revisited. Structural equation models: Present and future, A Festschrift in
honor of Karl Jöreskog, 2(3), pp.139-168.
38. Brandstätter, V., Lengfelder, A. and Gollwitzer, P.M., 2001, Implementation
intentions and efficient action initiation, Journal of personality and social
psychology, 81(5), p.946.
39. Brown, S.A. and Venkatesh, V., 2005, Model of adoption of technology in
households: A baseline model test and extension incorporating household
life cycle, MIS quarterly, pp.399-426.
40. Brown, S.A., Massey, A.P., Montoya-Weiss, M.M. and Burkman, J.R., 2002,
Do I really have to? User acceptance of mandated technology, European
journal of information systems, 11(4), pp.283-295.
41. Browne, R.H., 1995, On the use of a pilot sample for sample size
determination, Statistics in medicine, 14(17), pp.1933-1940.
86
42. Burton-Jones, A. and Hubona, G.S., 2006, The mediation of external

variables in the technology acceptance model, Information & management,
43(6), pp.706-717.
43. Byun, J.H., Park, M.H. and Jeong, B.Y., 2019, Effects of age and violations
on occupational accidents among motorcyclists performing food delivery,
Work, (Preprint), pp.1-9.
44. Cao, X., Yu, L., Liu, Z., Gong, M. and Adeel, L., 2018, Understanding
mobile payment users’ continuance intention: a trust transfer perspective,
Internet Research.
45. Chan, A.A.Y.H., Stahlman, W.D., Garlick, D., Fast, C.D., Blumstein, D.T.
and Blaisdell, A.P., 2010, Increased amplitude and duration of acoustic
stimuli enhance distraction, Animal behaviour, 80(6), pp.1075-1079.
46. Chan, F.K., Thong, J.Y., Venkatesh, V., Brown, S.A., Hu, P.J. and Tam,
K.Y., 2010, Modeling citizen satisfaction with mandatory adoption of an e-
government technology, Journal of the association for information systems,
11(10), pp.519-549.
47. Chen, C., Kang, Y., Huo, Z., Zhu, Z., Huang, W., Xin, H.L., Snyder, J.D., Li,
D., Herron, J.A., Mavrikakis, M. and Chi, M., 2014, Highly crystalline
multimetallic nanoframes with three-dimensional electrocatalytic surfaces,
Science, 343(6177), pp.1339-1343.
48. Cheng, E.W., 2001, SEM being more effective than multiple regression in
parsimonious model testing for management development research, Journal
of management development.
49. Chiu, C.M., Wang, E.T., Fang, Y.H. and Huang, H.Y., 2014, Understanding
customers' repeat purchase intentions in B2C e‐commerce: the roles of
utilitarian value, hedonic value and perceived risk, Information Systems
Journal, 24(1), pp.85-114.
50. Chong, A.Y.L., 2013, Predicting m-commerce adoption determinants: A
neural network approach, Expert Systems with Applications, 40(2), pp.523-
530.
87
51. Chopdar, P.K. and Sivakumar, V.J., 2019, Impulsiveness and its impact on
behavioural intention and use of mobile shopping apps: a mediation model,
International Journal of Business Innovation and Research, 19(1), pp.29-56.
52. Choudhary, N., 2019, Strategic Analysis of Cloud Kitchen-A Case Study,
Management Today, 9(3), pp.184-190.
53. Compeau, D.R. and Higgins, C.A., 1995, Computer self-efficacy:
Development of a measure and initial test, MIS quarterly, pp.189-211.
54. Cortina, J.M., 1993, What is coefficient alpha? An examination of theory and
applications, Journal of applied psychology, 78(1), p.98.
55. Crego, E.T. and Schiffrin, P.D., 1995, Customer-centered reengineering:
Remapping for total customer value, Irwin Professional Publishing.
56. Cronbach, L.J., 1951, Coefficient alpha and the internal structure of tests,
psychometrika, 16(3), pp.297-334.
57. Darke, P.R., Freedman, J.L. and Chaiken, S., 1995, Percentage discounts,
initial price, and bargain hunting: A heuristic-systematic approach to price
search behavior, Journal of Applied Psychology, 80(5), p.580.
58. Davis, F.D., 1989, Perceived usefulness, perceived ease of use, and user
acceptance of information technology, MIS quarterly, pp.319-340.
59. Davis, F.D., Bagozzi, R.P. and Warshaw, P.R., 1989, User acceptance of
computer technology: a comparison of two theoretical models, Management
science, 35(8), pp.982-1003.
60. Davis, F.D., Bagozzi, R.P. and Warshaw, P.R., 1992, Extrinsic and intrinsic
motivation to use computers in the workplace 1, Journal of applied social
psychology, 22(14), pp.1111-1132.
61. Dazmin, D. and Ho, M.Y., 2019, The relationship between consumers’ price-
saving orientation and time-saving orientation towards food delivery
intermediaries (FDI) services: an exploratory study, GSJ, 7(2).
62. De Hauw, S. and De Vos, A., 2010, Millennials’ career perspective and
psychological contract expectations: does the recession lead to lowered
expectations? Journal of business and psychology, 25(2), pp.293-302.
88
63. Deloitte, 2019, Future of Food: How technology & global trends are
transforming the food industry.
64. Deloitte, 2020, Impact of the COVID-19 crisis on consumer behavior.
65. DelVecchio, D. and Puligadda, S., 2012, The effects of lower prices on
perceptions of brand quality: a choice task perspective, Journal of Product
and Brand Management, 21(6), pp.465-474.
66. DeVellis, R.F., 2016, Scale development: Theory and applications, Sage
publications.
67. Dilshad, R.M. and Latif, M.I., 2013, Focus Group Interview as a Tool for
Qualitative Research: An Analysis, Pakistan Journal of Social Sciences
(PJSS), 33(1).
68. Ding, L., Velicer, W.F. and Harlow, L.L., 1995, Effects of estimation
methods, number of indicators per factor, and improper solutions on
structural equation modeling fit indices, Structural Equation Modeling: A
Multidisciplinary Journal, 2(2), pp.119-143.
69. Dlodlo, N., 2014, The relationships among service quality, trust, user
satisfaction and post-adoption intentions in M-payment services,
Mediterranean Journal of Social Sciences, 5(23), pp.165-165.
70. Dodds, W.B., Monroe, K.B. and Grewal, D., 1991, Effects of price, brand,
and store information on buyers’ product evaluations, Journal of marketing
research, 28(3), pp.307-319.
71. Dwivedi, Y.K., Rana, N.P., Janssen, M., Lal, B., Williams, M.D. and
Clement, M., 2017, An empirical validation of a unified model of electronic
government adoption (UMEGA), Government Information Quarterly, 34(2),
pp.211-230.
72. Dwivedi, Y.K., Rana, N.P., Jeyaraj, A., Clement, M. and Williams, M.D.,
2019, Re-examining the unified theory of acceptance and use of technology
(UTAUT): Towards a revised theoretical model, Information Systems
Frontiers, 21(3), pp.719-734.
89
73. Eriksson, K. and Nilsson, D., 2007, Determinants of the continued use of self-
service technology: The case of Internet banking, Technovation, 27(4),
pp.159-167.
74. Escobar-Rodríguez, T. and Carvajal-Trujillo, E., 2013, Online drivers of
consumer purchase of website airline tickets, Journal of Air Transport
75. Escobar-Rodríguez, T. and Carvajal-Trujillo, E., 2014, Online purchasing
tickets for low-cost carriers: An application of the unified theory of
acceptance and use of technology (UTAUT) model, Tourism Management,
43, pp.70-88.
76. Etikan, I., Musa, S.A. and Alkassim, R.S., 2016, Comparison of convenience
sampling and purposive sampling, American journal of theoretical and
applied statistics, 5(1), pp.1-4.
77. Facebook and Bain & Company, 2020, A SYNC SOUTHEAST ASIA
REPORT: Digital Consumers of Tomorrow, Here Today.
78. Facebook and Bain & Company, 2020, A SYNC SOUTHEAST ASIA
REPORT: Southeast Asia digital consumer trends that shape the next normal.
79. Fink, A., 2003, The survey handbook, Sage.
80. Fredricks, A.J. and Dossett, D.L., 1983, Attitude–behavior relations: A
comparison of the Fishbein-Ajzen and the Bentler-Speckart models, Journal
of personality and social psychology, 45(3), p.501.
81. Gallarza, M.G. and Saura, I.G., 2006, Value dimensions, perceived value,
satisfaction and loyalty: an investigation of university students’ travel
behaviour, Tourism management, 27(3), pp.437-452.
82. Gao, L., Waechter, K.A. and Bai, X., 2015, Understanding consumers’
continuance intention towards mobile purchase: A theoretical framework
and empirical study–A case of China, Computers in Human Behavior, 53,
pp.249-262.
83. Gaskin, J. and Lim, J., 2016, Master validity tool, AMOS Plugin In:
Gaskination’s StatWiki.
90
84. Gentry, L. and Calantone, R., 2002, A comparison of three models to explain
shop‐bot use on the web. Psychology & Marketing, 19(11), pp.945-956.
85. Goodchild, A. and Toy, J., 2018, Delivery by drone: An evaluation of
unmanned aerial vehicle technology in reducing CO2 emissions in the
delivery service industry, Transportation Research Part D: Transport and
Environment, 61, pp.58-67.
86. Goodhue, D.L. and Thompson, R.L., 1995, Task-technology fit and
individual performance, MIS quarterly, pp.213-236.
87. Gorsuch, R. (1983), Factor analysis, Hillsdale, NJ: L. Erlbaum Associates.
88. Greenwald, A.G. and Banaji, M.R., 1995, Implicit social cognition: attitudes,
self-esteem, and stereotypes, Psychological review, 102(1), p.4.
89. GSMA Intelligence, 2020, The Mobile Economy Asia Pacific 2020, GSM
Association.
90. Hair et al., Multivariate Data Analysis, 2010, 7th edition, Pearson Education
91. Hair Jr, J.F., Sarstedt, M., Hopkins, L. and Kuppelwieser, V.G., 2014, Partial
least squares structural equation modeling (PLS-SEM), European business
review.
92. Hair, J.F., Anderson, R.E., Tatham, R.L. and Black, W.C., 1999, Análisis
multivariante (Vol. 491), Madrid: Prentice Hall.
93. Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E. and Tatham, R.L., 2009,
Análise multivariada de dados, Bookman editora.
94. Hartman, J.L. and McCambridge, J., 2011, Optimizing millennials’
communication styles, Business Communication Quarterly, 74(1), pp.22-44.
95. Holbert, R.L. and Stephenson, M.T., 2002, Structural equation modeling in
the communication sciences, 1995–2000, Human Communication Research,
28(4), pp.531-551.
96. Hoogland, J.J. and Boomsma, A., 1998, Robustness studies in covariance
structure modeling: An overview and a meta-analysis, Sociological Methods
& Research, 26(3), pp.329-367.
91
97. Hoyle, R.H. and Kenny, D.A., 1999, Sample size, reliability, and tests of
statistical mediation, Statistical strategies for small sample research, 1,
pp.195-222.
98. Hoyle, R.H. ed., 1999, Statistical strategies for small sample research, Sage.
99. Hsiao, C.H., Chang, J.J. and Tang, K.Y., 2016, Exploring the influential
factors in continuance usage of mobile social Apps: Satisfaction, habit, and
customer value perspectives, Telematics and Informatics, 33(2), pp.342-355.
100. Hsiao, C.H., Chang, J.J. and Tang, K.Y., 2016, Exploring the influential
factors in continuance usage of mobile social Apps: Satisfaction, habit, and
customer value perspectives, Telematics and Informatics, 33(2), pp.342-355.
101. Hu, L.T. and Bentler, P.M., 1999, Cutoff criteria for fit indexes in covariance
structure analysis: Conventional criteria versus new alternatives, Structural
equation modeling: a multidisciplinary journal, 6(1), pp.1-55.
102. Hung, M.C., Yang, S.T. and Hsieh, T.C., 2012, An examination of the
determinants of mobile shopping continuance, International Journal of
Electronic Business Management, 10(1), p.29.
103. Hwang, J. and Choe, J.Y.J., 2019, Exploring perceived risk in building
successful drone food delivery services, International Journal of
Contemporary Hospitality Management.
104. Hwang, J. and Kim, H., 2019, Consequences of a green image of drone food
delivery services: The moderating role of gender and age, Business Strategy
and the Environment, 28(5), pp.872-884.
105. Iyer, P., Davari, A. and Mukherjee, A., 2018, Investigating the effectiveness
of retailers’ mobile applications in determining customer satisfaction and
repatronage intentions? A congruency perspective, Journal of Retailing and
Consumer Services, 44, pp.235-243.
106. Jensen, J.M., 2012, Shopping orientation and online travel shopping: The
role of travel experience, International Journal of Tourism Research, 14(1),
pp.56-70.
107. Jiang, Y., Wang, J., Wu, S., Li, N., Wang, Y., Liu, J., Xu, X., He, Z., Cheng,
Y., Zeng, X. and Wang, B., 2019, Association between Take-Out Food
92
Consumption and Obesity among Chinese University Students: A Cross-

Sectional Study, International journal of environmental research and public
health, 16(6), p.1071.
108. Kaewkitipong, L., Chen, C.C. and Ractham, P., 2016, A community-based
approach to sharing knowledge before, during, and after crisis events: A case
study from Thailand, Computers in Human Behavior, 54, pp.653-666.
109. Kang, J.W. and Namkung, Y., 2019, The role of personalization on
continuance intention in food service mobile apps, International Journal of
Contemporary Hospitality Management.
110. Karahanna, E., Straub, D.W. and Chervany, N.L., 1999, Information
technology adoption across time: a cross-sectional comparison of pre-
adoption and post-adoption beliefs, MIS quarterly, pp.183-213.
111. Khalifa, M., & Liu, V. (2004), The state of research on information system
satisfaction, Journal of Information Technology Theory and Application,
5(4), 37–49.
112. Khalilzadeh, J., Ozturk, A.B. and Bilgihan, A., 2017, Security-related factors
in extended UTAUT model for NFC based mobile payment in the restaurant
industry, Computers in Human Behavior, 70, pp.460-474.
113. Kim, S.S. and Malhotra, N.K., 2005, A longitudinal model of continued IS
use: An integrative view of four mechanisms underlying postadoption
phenomena, Management science, 51(5), pp.741-755.
114. Kline, T.J., 2005, Psychological testing: A practical approach to design and
evaluation, Sage Publications.
115. Koenig-Lewis, N., Marquet, M., Palmer, A. and Zhao, A.L., 2015, Enjoyment
and social influence: predicting mobile payment adoption, The Service
Industries Journal, 35(10), pp.537-554.
116. Konerding, U., 1999, Formal models for predicting behavioral intentions in
dichotomous choice situations, Methods of Psychological Research, 4(2),
pp.1-32.
117. Kurniawati, O.D., 2019, Analisis Faktor Adopsi Food Delivery Service App
Dengan Menggunakan Model Task Technology Fit (TTF), The Unified
93
Theory of Acceptance and Use Of Technology (UTAUT) Dan Initial Trust

Model (ITM), Universitas Airlangga.
118. Lai, I.K.W. and Shi, G., 2015, The impact of privacy concerns on the
intention for continued use of an integrated mobile instant messaging and
social network platform, International Journal of Mobile Communications,
13(6), pp.641-669.
119. Larsen, T.J., Sørebø, A.M. and Sørebø, Ø., 2009, The role of task-technology
fit as users’ motivation to continue information system use, Computers in
Human behavior, 25(3), pp.778-784.
120. Lee, K. and Joshi, K., 2007, An empirical investigation of customer
satisfaction with technology mediated service encounters in the context of
online shopping, Journal of information technology management, 18(2),
pp.18-37.
121. Lee, N.R. and Kotler, P., 2015, Social marketing: Changing behaviors for
good, Sage Publications.
122. Lee, S.W., Sung, H.J. and Jeon, H.M., 2019, Determinants of continuous
intention on food delivery apps: Extending UTAUT2 with information
quality, Sustainability (Switzerland), 11 (11), 3141.
123. Lee, Y. and Kwon, O., 2011, Intimacy, familiarity and continuance intention:
An extended expectation–confirmation model in web-based services,
Electronic Commerce Research and Applications, 10(3), pp.342-357.
124. Li, C., Mirosa, M. and Bremer, P., 2020, Review of Online Food Delivery
Platforms and their Impacts on Sustainability, Sustainability, 12(14), p.5528.
125. Liébana-Cabanillas, F., Alonso-Dos-Santos, M., Soto-Fuentes, Y. and
Valderrama-Palma, V.A., 2017, Unobserved heterogeneity and the
importance of customer loyalty in mobile banking, Technology Analysis &
Strategic Management, 29(9), pp.1015-1032.
126. Liébana-Cabanillas, F., Marinkovic, V., de Luna, I.R. and Kalinic, Z., 2018,
Predicting the determinants of mobile payment acceptance: A hybrid SEM-
neural network approach, Technological Forecasting and Social Change,
129, pp.117-130.
94
127. Likert, R., 1932, A technique for the measurement of attitudes, Archives of
psychology.
128. Limayem, M. and Cheung, C.M., 2008, Understanding information systems
continuance: The case of Internet-based learning technologies, Information
& management, 45(4), pp.227-232.
129. Limayem, M., Hirt, S.G. and Cheung, C.M., 2007, How habit limits the
predictive power of intention: The case of information systems continuance,
MIS quarterly, pp.705-737.
130. Lin, W.S., 2012, Perceived fit and satisfaction on web learning performance:
IS continuance intention and task-technology fit perspectives, International
Journal of Human-Computer Studies, 70(7), pp.498-507.
131. Liu, C. and Chen, J., 2019, Consuming takeaway food: Convenience, waste
and Chinese young people’s urban lifestyle, Journal of Consumer Culture,
p.1469540519882487.
132. Liu, C. and Chen, J., 2019. Consuming takeaway food: Convenience, waste
and Chinese
133. Lyytinen, K. and Hirschheim, R., 1988, Information systems failures—a
survey and classification of the empirical literature, In Oxford surveys in
information technology (pp. 257-309).
134. Madan, V. and Suri, R., 2001, Quality perception and monetary sacrifice: a
comparative analysis of discount and fixed prices, Journal of product &
brand management.
135. Maillet, É., Mathieu, L. and Sicotte, C., 2015, Modeling factors explaining
the acceptance, actual use and satisfaction of nurses using an Electronic
Patient Record in acute care settings: An extension of the UTAUT,
International journal of medical informatics, 84(1), pp.36-47.
136. Maillet, É., Mathieu, L. and Sicotte, C., 2015, Modeling factors explaining
the acceptance, actual use and satisfaction of nurses using an Electronic
Patient Record in acute care settings: An extension of the UTAUT,
International journal of medical informatics, 84(1), pp.36-47.
95
137. Maimaiti, M., Zhao, X., Jia, M., Ru, Y. and Zhu, S., 2018, How we eat
determines what we become: opportunities and challenges brought by food
delivery industry in a changing world in China, European journal of clinical
nutrition, 72(9), pp.1282-1286.
138. Marinković, V., Đorđević, A. and Kalinić, Z., 2020, The moderating effects
of gender on customer satisfaction and continuance intention in mobile
commerce: a UTAUT-based perspective, Technology Analysis & Strategic
Management, 32(3), pp.306-318.
139. Marsh, H.W. and Hau, K.T., 1999, Confirmatory factor analysis: Strategies
for small sample sizes, Statistical strategies for small sample research, 1,
pp.251-284.
140. Mathieson, K., 1991, Predicting user intentions: comparing the technology
acceptance model with the theory of planned behavior. Information systems
research, 2(3), pp.173-191.
141. McGill, T., Klobas, J. and Renzi, S., 2011, LMS use and instructor
performance: The role of task-technology fit, International Journal on E-
Learning, 10(1), pp.43-62.
142. Meah, A. and Jackson, P., 2017, Convenience as care: Culinary antinomies
in practice, Environment and Planning A, 49(9), pp.2065-2081.
143. Meenakshi, N. and Sinha, A., 2019, Food delivery apps in India: wherein
lies the success strategy, Strategic Direction.
144. Miniard, P.W. and Cohen, J.B., 1981, An examination of the Fishbein-Ajzen
behavioral-intentions model's concepts and measures, Journal of
Experimental Social Psychology, 17(3), pp.309-339.
145. Mintel, 2018, Asia Pacific The Food and Drink Landscape.
146. Moore, G.C. and Benbasat, I., 1991, Development of an instrument to
measure the perceptions of adopting an information technology innovation,
Information systems research, 2(3), pp.192-222.
147. Moreno, F.M., Lafuente, J.G., Carreón, F.Á. and Moreno, S.M., 2017, The
characterization of the millennials and their buying behavior, International
Journal of Marketing Studies, 9(5), pp.135-144.
96
148. Morning Consult, 2019, THE INFLUENCER REPORT: Engaging Gen Z and
Millennials.
149. Morosan, C. and DeFranco, A., 2016, It's about time: Revisiting UTAUT2 to
examine consumers’ intentions to use NFC mobile payments in hotels,
International Journal of Hospitality Management, 53, pp.17-29.
150. Muthén, B.O., 2002, Beyond SEM: General latent variable modeling,
Behaviormetrika, 29(1), pp.81-117.
151. Nakandala, D. and Lau, H.C., 2013, An application of a fuzzy-based
optimisation model for selecting food products based on cost and nutrition,
Journal of research for consumers, 24.
152. Nguyen, M., 2019, How Food Delivery Services in Vietnam Accommodate
Milennials and Generation Z: Case company: Delivery Now, Lahden
ammattikorkeakoulu.
153. Nunnally, J.C. and Bernstein, I.H., 1994, Psychological theory.
154. Nunnally, J.C., 1978, Psychometric Theory 2nd ed.
155. Okumus, B. and Bilgihan, A., 2014, Proposing a model to test smartphone
users' intention to use smart applications when ordering food in restaurants,
Journal of Hospitality and Tourism Technology.
156. Okumus, B., Ali, F., Bilgihan, A. and Ozturk, A.B., 2018, Psychological
factors influencing customers’ acceptance of smartphone diet apps when
ordering food at restaurants, International Journal of Hospitality
157. Oliveira, T., Faria, M., Thomas, M.A. and Popovič, A., 2014, Extending the
understanding of mobile banking adoption: When UTAUT meets TTF and
ITM, International journal of information management, 34(5), pp.689-703.
158. Oliver, R. L. (1980), A cognitive model for the antecedents and consequences
of satisfaction decisions, Journal of Marketing Research, 17, 460–469.
159. Palau-Saumell, R., Forgas-Coll, S., Sánchez-García, J. and Robres, E., 2019,
User acceptance of mobile apps for restaurants: An expanded and extended
UTAUT-2, Sustainability, 11(4), p.1210.
160. Pallant, J., 2007, SPSS survival manual, 3rd. Edition, McGrath Hill, 15.
97
161. Parasuraman, A., Zeithaml, V.A. and Berry, L.L., 1991, Refinement and
reassessment of the SERVQUAL scale, Journal of Retailing, Vol. 67 No. 4,
pp. 420-50.
162. Park, H.S., 2000, Relationships among attitudes and subjective norms:
Testing the theory of reasoned action across cultures, Communication
Studies, 51(2), pp.162-175.
163. Pillai, S.K.B., 2020, Customer continuance intention toward digital banking
applications. In Understanding Digital Industry: Proceedings of the
Conference on Managing Digital Industry, Technology and
Entrepreneurship (CoMDITE 2019), Bandung, Indonesia (p. 103).
164. Prensky, M., 2001, Digital natives, digital immigrants, On the horizon, 9(5).
165. Rainer, T.S. and Rainer, J., 2011, The millennials, B&H Publishing Group.
166. Raines, C., 2002, Managing millennials. Connecting Generations, The
Sourcebook, 16.
167. Raman, A. and Don, Y., 2013, Preservice teachers' acceptance of learning
management software: An application of the UTAUT2 model, International
Education Studies, 6(7), pp.157-164.
168. Raven, A., Leeds, E. and Park, C., 2010, Digital video presentation and
student performance: A task technology fit perspective, International Journal
of Information and Communication Technology Education (IJICTE), 6(1),
pp.17-29.
169. Ray, A., Dhir, A., Bala, P.K. and Kaur, P., 2019, Why do people use food
delivery apps (FDA)? A uses and gratification theory perspective, Journal of
Retailing and Consumer Services, 51, pp.221-230.
170. Research and Markets, 2020, Online Food Delivery Services Global Market
Report 2020-30: COVID-19 Growth and Change.
171. Rochon, J., Gondan, M. and Kieser, M., 2012, To test or not to test:
Preliminary assessment of normality when comparing two independent
samples, BMC medical research methodology, 12(1), p.81.
98
172. Roh, M. and Park, K., 2019, Adoption of O2O food delivery services in South
Korea: The moderating role of moral obligation in meal preparation,
International Journal of Information Management, 47, pp.262-273.
173. San Martín, H. and Herrero, Á., 2012, Influence of the user’s psychological
factors on the online purchase intention in rural tourism: Integrating
innovativeness to the UTAUT framework, Tourism Management, 33(2),
pp.341-350.
174. Schnettler, B., Rojas, J., Grunert, K.G., Lobos, G., Miranda-Zapata, E., Lapo,
M. and Hueche, C., 2019, Family and food variables that influence life
satisfaction of mother-father-adolescent triads in a South American country,
Current Psychology, pp.1-18.
175. Sekaran, S., 2003, Measurement: Scaling, reliability, validity, Research
methods for business: A skill building approach.
176. Shang, D. and Wu, W., 2017, Understanding mobile shopping consumers’
continuance intention, Industrial Management & Data Systems.
177. Shareef, M.A., Dwivedi, Y.K., Kumar, V. and Kumar, U., 2016, Reformation
of public service to meet citizens’ needs as customers: Evaluating SMS as an
alternative service delivery channel, Computers in Human Behavior, 61,
pp.255-270.
178. Sharif, A., Afshan, S. and Qureshi, M.A., 2019, Acceptance of learning
management system in university students: an integrating framework of
modified UTAUT2 and TTF theories, International Journal of Technology
Enhanced Learning, 11(2), pp.201-229.
179. Shaw, N. and Sergueeva, K., 2019, The non-monetary benefits of mobile
commerce: Extending UTAUT2 with perceived value, International Journal
of Information Management, 45, pp.44-55.
180. Shepard, G.J. and O’Keefe, D.J., 1984, Seperability of attitudinal and
normative influences on behavioral intentions in the Fishbein-Azjen model,
The Journal of Social Psychology, 122, pp.287-288.
181. Sheppard, B.H., Hartwick, J. and Warshaw, P.R., 1988, The theory of
reasoned action: A meta-analysis of past research with recommendations for
99
modifications and future research, Journal of consumer research, 15(3),

pp.325-343.
182. Shin, Y.M., Lee, S.C., Shin, B. and Lee, H.G., 2010, Examining influencing
factors of post-adoption usage of mobile internet: Focus on the user
perception of supplier-side attributes, Information Systems Frontiers, 12(5),
pp.595-606.
183. Song, G., Zhang, H., Duan, H. and Xu, M., 2018, Packaging waste from food
delivery in China’s mega cities, Resources, Conservation and Recycling,
130, pp.226-227.
184. Streiner, D.L., 2003, Being inconsistent about consistency: When coefficient
alpha does and doesn't matter, Journal of personality assessment, 80(3),
pp.217-222.
185. Sun, J., & Chi, T., 2018, Key factors influencing the adoption of apparel
mobile commerce: An empirical study of Chinese consumers, The Journal of
the Textile Institute.
186. Susanto, A., Chang, Y. and Ha, Y., 2016, Determinants of continuance
intention to use the smartphone banking services, Industrial Management &
Data Systems.
187. Tabachnick, B.G. and Fidell, L.S., 2001, Using multivariate statistics, Allyn
and Bacon, Needham Heights, MA.
188. Tabachnick, B.G. and, L.S., 2001, Using multivariate statistics, Allyn and
Bacon, Needham Heights, MA.
189. Tam, C. and Oliveira, T., 2016, Understanding the impact of m-banking on
individual performance: DeLone & McLean and TTF perspective,
Computers in Human Behavior, 61, pp.233-244.
190. Tam, C., Santos, D. and Oliveira, T., 2020, Exploring the influential factors
of continuance intention to use mobile Apps: Extending the expectation
confirmation model, Information Systems Frontiers, 22(1), pp.243-257.
191. Tamam, E., Hassan, M.S.H., Waheed, M. and Zaremohzzabieh, Z., 2016,
Factors affecting Malaysian university students’ purchase intention in social
networking sites, Cogent Business & Management, 3(1), p.1182612.
100
192. Tamilmani, K., Rana, N.P., Dwivedi, Y., Sahu, G.P. and Roderick, S., 2018,
Exploring the Role of'Price Value'for Understanding Consumer Adoption of
Technology: A Review and Meta-analysis of UTAUT2 based Empirical
Studies, PACIS, p. 64.
193. Tan, A., Xiao, M., Cui, X., Chen, X., Chen, Y., Fang, D., Fu, C., Giboni, K.,
Giuliani, F., Gong, H. and Guo, X., 2016, Dark matter results from first 98.7
days of data from the PandaX-II experiment, Physical review letters, 117(12),
p.121303.
194. Taylor, P. and Keeter, S., 2010, MILLENNIALS: A Portrait of Generation
Next. Confident. Connected. Open to Change, Pew Research Center.
195. Taylor, S. and Todd, P.A., 1995, Understanding information technology
usage: A test of competing models, Information systems research, 6(2),
pp.144-176.
196. Thaler, R.H., 2008, Mental accounting and consumer choice, Marketing
Science, 27(1), pp.15-25.
197. Thominathan, S. and Ramayah, T., 2015, Ensuring continued usage of an E-
Government service in Malaysia: the role of perceived usefulness and user
satisfaction, Public affairs and administration: concepts, methodologies,
tools, and applications, pp. 1546-1562, IGI Global.
198. Tinsley, H.E. and Tinsley, D.J., 1987, Uses of factor analysis in counseling
psychology research, Journal of counseling psychology, 34(4), p.414.
199. Troise, C., O'Driscoll, A., Tani, M. and Prisco, A., 2020, Online food delivery
services and behavioural intention–a test of an integrated TAM and TPB
framework, British Food Journal.
200. Tse, D.K. and Wilton, P.C., 1988, Models of consumer satisfaction
formation: An extension, Journal of marketing research, 25(2), pp.204-212.
201. Uhlmann, E. and Swanson, J., 2004, Exposure to violent video games
increases automatic aggressiveness, Journal of adolescence, 27(1), pp.41-52.
202. Van der Heijden, H., 2004, User acceptance of hedonic information systems,
MIS quarterly, pp.695-704.
101
203. Venkatesh, V., Morris, M.G., Davis, G.B. and Davis, F.D., 2003, User
acceptance of information technology: Toward a unified view, MIS quarterly,
pp.425-478.
204. Venkatesh, V., Thong, J.Y. and Xu, X., 2012, Consumer acceptance and use
of information technology: extending the unified theory of acceptance and
use of technology, MIS quarterly, pp.157-178.
205. Verkijika, S.F., 2018, Factors influencing the adoption of mobile commerce
applications in Cameroon, Telematics and Informatics, 35(6), pp.1665-1674.
206. Wangpipatwong, S., Chutimaskul, W. and Papasratorn, B., 2008,
Understanding Citizen's Continuance Intention to Use e-Government
Website: a Composite View of Technology Acceptance Model and Computer
Self-Efficacy, Electronic Journal of e-Government, 6(1).
207. Warshaw, P.R., 1980, A new model for predicting behavioral intentions: An
alternative to Fishbein, Journal of marketing research, 17(2), pp.153-172.
208. Wegner, D.M. and Wheatley, T., 1999, Apparent mental causation: Sources
of the experience of will. American psychologist, 54(7), p.480.
209. Wegner, D.M., 2002, The illusion of conscious will, MIT press.
210. Wen, I., 2012, An empirical study of an online travel purchase intention
model, Journal of Travel & Tourism Marketing, 29(1), pp.18-39.
211. Wen, Z., Zhang, Y. and Fu, D., 2019, The environmental impact assessment
of a takeaway food delivery order based on of industry chain evaluation in
China. China Environ, Sci, 39, pp.4017-4024.
212. Westland, J.C., 2010, Lower bounds on sample size in structural equation
modeling, Electronic commerce research and applications, 9(6), pp.476-487.
213. Wolf, E.J., Harrington, K.M., Clark, S.L. and Miller, M.W., 2013, Sample
size requirements for structural equation models: An evaluation of power,
bias, and solution propriety, Educational and psychological measurement,
73(6), pp.913-934.
214. Wu, B. and Chen, X., 2017, Continuance intention to use MOOCs:
Integrating the technology acceptance model (TAM) and task technology fit
(TTF) model, Computers in Human Behavior, 67, pp.221-232.
102
215. Yang, H.D. and Yoo, Y., 2004, It's all about attitude: revisiting the
technology acceptance model, Decision support systems, 38(1), pp.19-31.
216. Yeh, Y.S. and Li, Y.M., 2009, Building trust in m‐commerce: contributions
from quality and satisfaction, Online Information Review.
217. Yen, D.C., Wu, C.S., Cheng, F.F. and Huang, Y.W., 2010, Determinants of
users’ intention to adopt wireless technology: An empirical study by
integrating TTF with TAM, Computers in Human Behavior, 26(5), pp.906-
915.
218. Yeo, V.C.S., Goh, S.K. and Rezaei, S., 2017, Consumer experiences, attitude
and behavioral intention toward online food delivery (OFD) services,
Journal of Retailing and Consumer Services, 35, pp.150-162.
219. Yockey, R.D., 2016. SPSS demystified: A simple guide and reference,
Routledge.
220. Yuan, S., Liu, Y., Yao, R. and Liu, J., 2016, An investigation of users’
continuance intention towards mobile banking in China, Information
Development, 32(1), pp.20-34.
221. Zhang, Y., Huang, Y., Wang, Y. and Casey, T.W., 2020, Who uses mobile
phone while driving for food delivery? The role of personality, risk
perception, and driving self-efficacy, Journal of Safety Research.
222. Zhao, Y. and Bacao, F., 2020, What factors determining customer
continuingly using food delivery apps during 2019 novel coronavirus
pandemic period, International journal of hospitality management, 91,
p.102683.
223. Zhou, T. and Li, H., 2014, Understanding mobile SNS continuance usage in
China from the perspectives of social influence and privacy concern,
Computers in Human Behavior, 37, pp.283-289.
III. References from the Internet
1. McKinsey & Company, 2016, The changing market for food delivery,
available at https://www.mckinsey.com/industries/technology-media-and-
103
telecommunications/our-insights/the-changing-market-for-food-delivery
[accessed 20/10/2020].
2. McKinsey & Company, 2017, Cracking the code on millennial consumers,
available at https://www.mckinsey.com/industries/retail/our-
insights/cracking-the-code-on-millennial-consumers [accessed
20/10/2020].
3. Statista, 2020, Apps (Viet Nam), available at
https://www.statista.com/outlook/318/127/apps/vietnam [accessed
20/10/2020].
4. Statista, 2020, Frequency of ordering food on food delivery apps in Vietnam
as of June 2020, by age group, available at
https://www.statista.com/statistics/1153968/vietnam-ordering-food-from-
food-delivery-apps-frequency-by-age-group/ [accessed 20/10/2020].
5. Statista, 2020, Online Food Delivery (Viet Nam), available at
https://www.statista.com/outlook/374/127/online-food-delivery/vietnam
[accessed 20/10/2020].
6. Statista, 2020, Share of respondents who would keep using food delivery
apps as often after social distancing measures were lifted due to COVID-19
pandemic in Vietnam as of June 2020, available at
https://www.statista.com/statistics/1154084/vietnam-food-delivery-apps-
usage-post-covid-19/ [accessed 20/10/2020].
7. Think with Google, 2018, The Impatient Consumer: Making decisions faster
than ever, available at https://www.thinkwithgoogle.com/consumer-
insights/consumer-trends/managingconsumerdemands/ [accessed
20/10/2020].
8. Vietcetera, 2020, A Brand’s Story: BAEMIN Builds a Culture of
Conversationalists, available at https://vietcetera.com/en/a-brands-story-
baemin-builds-a-culture-of-conversationists [accessed 20/10/2020]
9. Vietnam Investment Review, 2019, Delivery platforms beckon food stalls,
available at https://www.vir.com.vn/delivery-platforms-beckon-food-stalls-
68746.html [accessed 20/10/2020]
104
10. Vietnam Investment Review, 2020, Food-hailing action nothing but a drop
in ocean of plastic, available at https://www.vir.com.vn/food-hailing-action-
nothing-but-a-drop-in-ocean-of-plastic-78746.html [accessed 20/10/2020]
11. Vietnam Investment Review, 2020, Grab shakes up F&B market with cloud
kitchen concept, available at https://www.vir.com.vn/grab-shakes-up-fb-
market-with-cloud-kitchen-concept-71565.html [accessed 20/10/2020]
105
APPENDICES
Appendix 1: The market outlook of online delivery segment in major

countries ............................................................................................................. 108
Appendix 2: Performance of PCD & RCD in terms of user number, revenue

growth & average revenue per user ................................................................ 109
Appendix 3: Survey findings by Q&Me on 840 respondents in HCM city and

Hanoi in April 2020 ........................................................................................... 110
Appendix 3.1: Food ordering frequency ........................................................ 110
Appendix 3.2: Food ordering occasions ........................................................ 111
Appendix 3.3: Reasons to like food delivery apps ........................................ 111
Appendix 4: Birth year range of the Millennial generation .......................... 112
Appendix 5: Theory of Reasoned Action (TRA) (1975) ................................ 113
Appendix 6: Technology Acceptance Model (TAM) (1989) .......................... 114
Appendix 7: Unified Theory of Acceptance and Use of Technology (UTAUT)

(2003) .................................................................................................................. 115
Appendix 8: Some relevant studies conducted on intention to use or related

topics ................................................................................................................... 116
Appendix 9: List of experts within the Marketing and Communication fields

............................................................................................................................. 118
Appendix 10: Outline for focus group interview............................................ 119
Appendix 11: Result of focus group interview ............................................... 125
Appendix 12: Cronbach’s Alpha coefficients for pilot study ........................ 129
Appendix 13: Adjusted measurement scales .................................................. 137
Appendix 14: Official questionnaire ............................................................... 140
Appendix 15: Descriptive statistics analysis by demographics factors ........ 148

106
Appendix 15.1: Summary of descriptive statistics analysis by demographics

factors ............................................................................................................. 148
Appendix 15.2: Descriptive statistics analysis by demographics factors ...... 149
Appendix 16: Descriptive statistics analysis by measurement scales ...........151
Appendix 17: Cronbach’s Alpha coefficients for official study ....................153
Appendix 17.1: Summary of Cronbach’s Alpha coefficients for official study

........................................................................................................................ 153
Appendix 17.2: Cronbach’s Alpha coefficients for official study ................. 155
Appendix 18: Explanatory Factor Analysis (EFA) ........................................163
Appendix 18.1: Explanatory Factor Analysis (EFA) for 10 variables affecting

Continuance Intention (CI)............................................................................. 163
Appendix 18.2: Explanatory Factor Analysis (EFA) for Continuance Intention

(CI) ................................................................................................................. 170
Appendix 19: Confirmatory Factor Analysis (CFA) .....................................172
Appendix 19.1: CFA path diagram ................................................................ 172
Appendix 19.2: Model-fit indices of measurement model............................. 172
Appendix 19.3: Standardized Regression Weights ........................................ 176
Appendix 19.4: Validity and reliability test ................................................... 179
Appendix 20: Structural Equation Modeling (SEM) .....................................180
Appendix 20.1: SEM path diagram ................................................................ 180
Appendix 20.2: Indices for model fit, validity & reliability .......................... 181
Appendix 20.3: Regression Weights .............................................................. 187
Appendix 20.4: Standardized Regression Weights ........................................ 188
Appendix 20.5: Squared Multiple Correlations ............................................. 191
Appendix 21: Independent Sample t-Test & One-Way ANOVA on

demographic variables ......................................................................................194
Appendix 21.1: Gender .................................................................................. 194

107
Appendix 21.2: Age ....................................................................................... 195
Appendix 21.3: Income .................................................................................. 196
Appendix 21.4: Educational background ....................................................... 197

108
Appendix 1: The market outlook of online delivery segment in major countries
Volume of market’s
Forecast Annual growth Market’s
largest delivery
No. Country revenue in 2020 rate (CAGR1 largest delivery
segment in 2020
(million USD) 2020–2024) (%) segment
(million USD)
1 China 51,514 7.0 PCD 37,708
The
2 United 26,527 5.1 RCD 15,631
States
3 India 10,196 9.5 RCD 5,401
The
United
4 5,988 6.5 RCD 4,115
Kingdom
Brazil
5 3,300 9.5 RCD 2,033
1
CAGR stands for Compound Annual Growth Rate, which is the mean annual growth rate of a business
segment or an investment over a specified period of time longer than one year.
109
Appendix 2: Performance of PCD & RCD in terms of user number, revenue

growth & average revenue per user2
Year 2017 2018 2019 2020 2021 2022 2023 2024
User number PCD 0.6 0.9 1.4 2.2 2.9 3.5 4.1 4.6
(million)
RCD 3.7 4.9 6.2 8.3 9.7 10.9 12.0 12.8
Revenue PCD N/A 78.2 62.7 64.2 33.3 23.9 17.2 12.4
growth
RCD N/A 38.0 34.1 39.9 21.5 16.1 11.9 8.7
(Percent)
Average PCD 31.10 33.66 35.81 37.41 38.50 39.21 39.66 39.95
revenue per
RCD 22.71 23.82 25.05 26.24 27.28 28.12 28.76 29.23
user (USD)
2
Table data from 2020 onwards are estimates only.
110
Appendix 3: Survey findings by Q&Me on 840 respondents in HCM city and

Hanoi in April 2020
Appendix 3.1: Food ordering frequency
More than once a day 6%
Everyday 7%
5-6 times a week 14%
3-4 times a week 26%
Once or twice a week 26%
2-3 times a month 16%
Once a month 3%
Less than once a month 2%
Percentage of respondents
Source: Q&Me, 2020

111
Appendix 3.2: Food ordering occasions
Lunch 51%
Teatime 51%
Dinner 43%
When I feel hungry 26%
Breakfast 16%
Supper 11%
Only if promotions 8%
Others 1%
Percentage of respondents
Source: Q&Me, 2020
Appendix 3.3: Reasons to like food delivery apps
Promotion 11%
Good price 11%
Good attitude 11%
Convenience 14%
Fast delivery 30%
Good services 31%
Percentage of satisfied respondents
Source: Q&Me, 2020

112
Appendix 4: Birth year range of the Millennial generation
No. Institute/Author (Year) Birth year range
1 Valentine & Powers (2013) 1977 - 1996
2 Kotler & Armstrong (2018) 1977-2000
3 Muda, Mohd, & Hassan (2016) 1980's to the early 1990's
4 Lissitsa & Kol (2016) 1980-1999
5 Omar et al (2016) 1980-1990
6 Pew Research Center (2016) 1981-1996
8 Junker, Walcher, & Blazek (2016) 1981-1995
9 Ordun (2015) 1981-2000
10 Howe & Strauss (2000) 1982-1988
Source: Self-deprived from author’s literature review,

2020
113
Appendix 5: Theory of Reasoned Action (TRA) (1975)
Source: Fishbein and Ajzen, 1975
Attitude toward the behavior refers to one’s general feeling of favorableness

for performing such behavior.
Subjective Norm can be understood as to one’s perception that most people

who are important to him think he should or should not engage in certain behavior.
114
Appendix 6: Technology Acceptance Model (TAM) (1989)
Source: Davis, 1986

115
Appendix 7: Unified Theory of Acceptance and Use of Technology (UTAUT)

(2003)
Source: Venkatesh et al., 2003

116
Appendix 8: Some relevant studies conducted on intention to use or related

topics
Research Authors Framework Variables Finding Gap(s)
Why do Ray et al. U &G theory convenience, Out of eight proposed Geographically restricted
people use societal pressure, variables, customer in India
food delivery customer experience, search of
apps (FDA)? experience, delivery restaurants, ease-of-use
A uses and experience, search and listing were proven
gratification of restaurants, significant antecedents of
theory quality control, intentions to use FDAs
perspective listing, and ease-of-
(2019) use
User Palau- Extended Effort expectancy, All given drivers are (1) Self-reported
Acceptance of Saumell UTAUT-2 performance significantly supported, measurement of use, not
Mobile Apps et al. with expectancy, with habit being the actual use data
for perceived hedonic motivation, strongest determinant of
(2) Geographically
Restaurants: credibility facilitating intention to use, and of
restricted to Spain
An Expanded conditions, price- actual usage
and Extended saving orientation, (3) Not including tourist,
UTAUT-2 habit, social an important MARSR
(2019) influence, and user group in Spain.
perceived
credibility
Consumer Yeo et al. Contingency Convenience All proposed factors (1) Limited number of
experience, Framework motivation, post- significantly support responses (224)
attitude and & Extended usage usefulness, attitude and behavioral (2) Mostly Chinese
behavioral Model of IT hedonic motivation, intention towards OFD ethnicity students (18–22
intention Continuance price saving services. Yet, the years old) -insufficient to
toward online orientation, time relationship between be a representation of the
food delivery saving orientation, prior online purchase population
(OFD) prior online experience and post-
(3) Geographically
services purchase experience usage usefulness wasn’t
restricted to the Klang
(2017) supported
Valley, Malaysia
117
Consumer’ Gunden UTAUT2 performance Except impulse buying (1) solely focusing on the
intentions to et al. and three expectancy, habit, tendency, other variables behaviors of consumers
use online additional impulse buying are statistically vis-à-vis restaurants in
food delivery constructs tendency, congruity supported, with the USA
systems in the with self-image and performance expectancy
(2) a comprehensive yet
USA (2020) mindfulness being the strongest
parsimonious conceptual
predictor
model
Millennial Suhartant E-S-QUAL e-service quality All are critical (1) not differentiating in-
experience o et al. model (E- (privacy, efficiency, determinants of house and third-party
with online service system availability, millennial satisfaction, delivery services
food home quality) with fulfillment), food with e-service quality
(2) One construct had a
delivery: A food quality quality and being higher than the
factor loading value
lesson from and perceived value others
below suggested level,
Indonesia perceived
indicating one indicator
(2019) value
has low statistical
validity
(3) Geographically
restricted in Indonesia
118
Appendix 9: List of experts within the Marketing and Communication fields
No. Name Tittle Email

1 Pham President, EnvZone (a linh.pham@envzone.com
Linh Colorado-based
outsourcing consultancy
company)
2 Vu Thi Scientific Engagement & thithanhthao.vu@vpmbr.com.vn
Thanh Communications
Thao, Manager, Philip Morris
MD, International (Vietnam
PhD, Branch)
3 Thai Communications dangngocchau.thai@vpmbr.com.vn
Dang Manager, Philip Morris
Ngoc International (Vietnam
Chau Branch)
4 Le Giang Marketing Lecturer, legiangnam.cs2@ftu.edu.vn
Nam, Foreign Trade University
MBA
119
Appendix 10: Outline for focus group interview
Xin chào Anh/Chị,

Tôi tên là Nguyễn Thị Thuận An, hiện đang là sinh viên năm cuối trường Đại
học Ngoại thương cơ sở II tại TP.HCM. Hiện tại tôi đang thực hiện khóa luận tốt
nghiệp với đề tài “Nghiên cứu các nhân tố ảnh hưởng đến ý định tiếp tục sử dụng
các ứng dụng giao thức ăn trực tuyến của thế hệ Millennial tại khu vực TP.HCM”
và rất mong nhận được sự hợp tác của Anh/Chị trong việc hoàn thành nghiên cứu
này.
Không có quan điểm nào đúng hay sai, vì vậy tất cả ý kiến của Anh/Chị đều
là các thông tin hữu ích cho nghiên cứu. Vì vậy, tôi rất mong Anh/Chị dành một chút
thời gian để hoàn thành bảng phỏng vấn dưới đây.
1. Câu hỏi về các yếu tố ảnh hưởng trực tiếp và/hoặc gián tiếp đến ý định tiếp
tục sử dụng các ứng dụng giao thức ăn trực tuyến của thế hệ Millennial tại khu
vực TP.HCM:
Theo Anh/Chị, những yếu tố từ (1) đến (7) và yếu tố (10) có ảnh hưởng đến ý
định tiếp tục sử dụng các ứng dụng giao thức ăn trực tuyến dưới góc độ cảm nhận
của Anh/Chị hay không? Vì sao?
Theo Anh/Chị, những yếu tố từ (1) đến (5), yếu tố (7) và yếu tố (9) có ảnh
hưởng đến mức độ hài lòng khi sử dụng các ứng dụng giao thức ăn trực tuyến dưới
góc độ cảm nhận của Anh/Chị hay không? Vì sao?
Theo Anh/Chị, những yếu tố (2), (8), (9) có ảnh hưởng đến hiệu quả mong
đợi khi sử dụng các ứng dụng giao thức ăn trực tuyến dưới góc độ cảm nhận của
Anh/Chị hay không? Vì sao?
STT Yếu tố Giải thích yếu tố
Các lợi ích và tiện ích (chẳng hạn như: tiết kiệm thời gian,
Hiệu quả
1 công sức, sự làm việc hiệu quả, khả năng tiếp cận, khả năng
mong đợi
tùy chỉnh, sự tiện lợi) mà người dùng có thể có được, nhằm
120
thực hiện một mục đích nào đó, khi sử dụng một hệ thống
công nghệ
Nỗ lực kỳ
2 Mức độ dễ dàng khi sử dụng một hệ thống công nghệ
vọng
Mức độ đón nhận của một cá nhân khi những người quan trọng
Ảnh hưởng
3 với cá nhân đó (người thân, bạn bè, đồng nghiệp, etc) nghĩ
xã hội
rằng anh hoặc chị ta nên sử dụng một hệ thống mới
Mức độ một cá nhân tin rằng họ sẽ nhận được những sự hỗ trợ

Điều kiện hỗ
4 về mặt kỹ thuật cũng như từ đội ngũ nhân viên khi sử dụng hệ
trợ
thống
Động lực thụ

5 Niềm vui hay sự thú vị có được từ việc sử dụng công nghệ
hưởng
Mức độ mà một cá nhân có xu hướng thực hiện hành vi một

6 Thói quen
cách tự động sau một thời gian tích lũy kiến thức và kĩ năng.
Sự có sẵn của các mã giảm giá, ưu đãi, chương trình khuyến

Xu hướng
7 mãi đặc biệt, và/hoặc khả năng so sánh giá thông qua những
tiết kiệm
thông tin được cung cấp về giá thành và sản phẩm
Sự phù hợp
giữa công Sự tương thích giữa các tính năng của một hệ thống công nghệ
8
nghệ và công với công việc/nhu cầu của người dùng
việc
Sự đáp ứng Mức độ việc trải nghiệm một hệ thống công nghệ thực sự thỏa
9
kỳ vọng mãn (những) kỳ vọng ban đầu của người dùng về hệ thống này
Kết quả của một quá trình so sánh giữa kỳ vọng và hiệu năng
10 Sự hài lòng thực tế của hệ thống (Nói cách khác, đây là kết quả của các xác
nhận hoặc không xác nhận tích cực của sự mong đợi, có nghĩa
121
là hiệu suất thực tế thấp hơn/tương đương/tốt hơn so với kết

quả dự kiến)
2. Câu hỏi về các biến quan sát trong thang đo các yếu tố ảnh hưởng trực tiếp
và/hoặc gián tiếp đến ý định tiếp tục sử dụng các ứng dụng giao thức ăn trực
tuyến của thế hệ Millennial tại khu vực TP.HCM:
Đối với từng thang đo dưới đây, các phát biểu được cho là thể hiện suy nghĩ
và sự quan tâm của Anh/Chị về ý định tiếp tục sử dụng của Anh/Chị đối với các ứng
dụng giao thức ăn trực tuyến. Xin Anh/Chị vui lòng đọc kỹ, sau đó trả lời các câu hỏi
sau:
(1) Anh/Chị có đồng ý với các phát biểu hoặc có góp ý bổ sung, chỉnh sửa gì
đối với các phát biểu không? Vì sao?
(2) Ngoài ra, Anh/Chị có hiểu rõ các phát biểu không? Nếu không, xin
Anh/Chị vui lòng đóng góp ý kiến về nội dung sao cho dễ hiểu hơn.
1. Hiệu quả mong đợi của các ứng dụng giao thức ăn trực tuyến (4 phát biểu):
Tôi thấy các ứng dụng giao thức ăn trực tuyến có ích cho việc đặt và nhận thức ăn của
tôi
Tôi thấy các ứng dụng giao thức ăn trực tuyến tiện lợi để đặt và nhận thức ăn của tôi
Tôi thấy các ứng dụng giao thức ăn trực tuyến cải tiến quy trình đặt và nhận thức ăn
của tôi
Tôi thấy các ứng dụng giao thức ăn trực tuyến nâng cao hiệu quả việc đặt và nhận
thức ăn của tôi
2. Nỗ lực kỳ vọng khi sử dụng các ứng dụng giao thức ăn trực tuyến (3 phát biểu):
Tôi thấy mình có thể dễ dàng học cách sử dụng các ứng dụng giao thức ăn trực tuyến
Tôi thấy các tương tác khi sử dụng các ứng dụng giao thức ăn trực tuyến rất rõ ràng và
dễ hiểu
122
Tôi thấy các bước đặt thức ăn để giao trên các ứng dụng giao thức ăn trực tuyến rất dễ
để làm theo
Tôi thấy dễ dàng để sử dụng các ứng dụng giao thức ăn trực tuyến một cách thành
thục
3. Ảnh hưởng xã hội lên quyết định (tiếp tục) sử dụng các ứng dụng giao thức ăn
trực tuyến (4 phát biểu):
Những người quan trọng với tôi (gia đình, bạn bè, đồng nghiệp, Influencer, v.v.) đề
xuất tôi (tiếp tục) sử dụng ứng dụng giao thức ăn trực tuyến
Những người quan trọng với tôi nghĩ rằng ứng dụng giao thức ăn trực tuyến mang lợi
ích cho tôi
Những người quan trọng với tôi nghĩ rằng việc (tiếp tục) sử dụng ứng dụng giao thức
ăn trực tuyến là một ý kiến hay
Những người quan trọng với tôi ủng hộ tôi trong việc (tiếp tục) sử dụng ứng dụng giao
thức ăn trực tuyến
4. Điều kiện hỗ trợ khi sử dụng các ứng dụng giao thức ăn trực tuyến (4 phát biểu):
Tôi có điện thoại thông minh để sử dụng ứng dụng giao thức ăn trực tuyến
Tôi thấy tôi có đủ kiến thức để sử dụng ứng dụng giao thức ăn trực tuyến
Tôi thấy ứng dụng giao thức ăn trực tuyến tương thích với những công nghệ khác mà
tôi sử dụng
Tôi thấy tôi hoàn toàn thoải mái khi sử dụng ứng dụng giao thức ăn trực tuyến
5. Động lực thụ hưởng khi sử dụng các ứng dụng giao thức ăn trực tuyến (3 phát
biểu);
Tôi thấy việc sử dụng ứng dụng giao thức ăn trực tuyến vui
Tôi thấy việc sử dụng ứng dụng giao thức ăn trực tuyến thú vị
Tôi tận hưởng trải nghiệm sử dụng các ứng dụng giao thức ăn trực tuyến
123
6. Xu hướng tiết kiệm khi sử dụng các ứng dụng giao thức ăn trực tuyến (3 phát
biểu):
Tôi có thể tiết kiệm tiền khi được cung cấp thông tin và so sánh giá giữa các cửa hàng
khác nhau trên ứng dụng giao thức ăn trực tuyến
Tôi thích việc tìm kiếm và sử dụng các ưu đãi khi sử dụng ứng dụng giao thức ăn trực
tuyến
Tôi thấy trong cùng một khoản chi phí, việc sử dụng ứng dụng giao thức ăn trực tuyến
mang đến những giá trị tiêu dùng tốt hơn cho tôi
7. Thói quen sử dụng các ứng dụng giao thức ăn trực tuyến (4 phát biểu):
Việc sử dụng ứng dụng giao thức ăn trực tuyến đã trở thành thói quen đối với tôi
Tôi thích sử dụng ứng dụng giao thức ăn trực tuyến
Tôi thấy tôi cần sử dụng ứng dụng giao thức ăn trực tuyến
Việc sử dụng ứng dụng giao thức ăn trực tuyến trở nên rất bình thường đối với tôi
8. Sự phù hợp giữa công nghệ và công việc của các ứng dụng giao thức ăn trực
tuyến (3 phát biểu):
Các tính năng trên ứng dụng giao thức ăn trực tuyến đủ để tôi đặt và nhận thức ăn
Các tính năng trên ứng dụng giao thức ăn trực tuyến phù hợp để tôi kiểm tra quy trình
đặt và nhận thức ăn
Các tính năng trên ứng dụng giao thức ăn trực tuyến đáp ứng đầy đủ các yêu cầu của
tôi về đặt và nhận thức ăn
9. Sự đáp ứng kỳ vọng của tôi sau khi trải nghiệm sử dụng các ứng dụng giao thức
ăn trực tuyến (4 phát biểu):
Trải nghiệm sử dụng ứng dụng giao thức ăn trực tuyến tốt hơn tôi mong đợi
Các tính năng trên ứng dụng giao thức ăn trực tuyến nhiều hơn những gì tôi mong đợi
124
Dịch vụ được cung cấp bởi ứng dụng giao thức ăn trực tuyến tốt hơn tôi mong đợi
Nhìn chung, hầu hết những mong đợi của tôi khi sử dụng ứng dụng giao thức ăn trực
tuyến được đáp ứng
10. Mức độ hài lòng của tôi khi sử dụng các ứng dụng giao thức ăn trực tuyến (4
phát biểu):
Tôi cảm thấy hài lòng khi các ứng dụng giao thức ăn trực tuyến đáp ứng những yêu
cầu của tôi
Tôi cảm thấy hài lòng với sự hiệu quả của các ứng dụng giao thức ăn trực tuyến
Tương tác của tôi với các ứng dụng giao thức ăn trực tuyến làm tôi hài lòng
Tôi tin rằng quyết định sử dụng các ứng dụng giao thức ăn trực tuyến đã là điều đúng
đắn
11. Ý định tiếp tục sử dụng các ứng dụng giao dụng giao thức ăn trực tuyến (4 phát
biểu):
Tôi định sẽ tiếp tục sử dụng các ứng dụng giao thức ăn trực tuyến trong tương lai
Tôi sẽ luôn cố gắng sử dụng các ứng dụng giao thức ăn trực tuyến trong cuộc sống
hàng ngày
Tôi có kế hoạch sử dụng các ứng dụng giao thức ăn trực tuyến thường xuyên
Tôi đã quyết định sẽ sử dụng các ứng dụng giao thức ăn trực tuyến để đặt thức ăn cho
lần tiếp theo
Xin chân thành cảm ơn sự hợp tác của Qúy Anh/Chị!

125
Appendix 11: Result of focus group interview

1. List of interviewees
FDA usage pattern

Frequency
No. Name Age Job
(ordering times Specific FDAs
per month)
1 Luong Minh Chau 21 12 – below 20
2 Trinh Gia Han 21 20 and above
University Now, Baemin,
3 Doan Tan Khang 21 20 and above
student GrabFood, GoFood
4 Nguyen Thi Kieu Oanh 21 6 – below 12
5 Bach Vo Phuong Vy 20 12 – below 20
6 Vu Anh Tu 23 12 – below 20 Now, Baemin
Tran Hoang Khanh GrabFood,
7 24 20 and above
Ngoc GoFood, Baemin
First Jobber
8 Tran Doan Thuc 24 20 and above Baemin
Doan Nguyen Thanh GrabFood, Now,
9 24 12 – below 20
Tam Baemin
10 Pham Thi Nho 31 White- 12 – below 20 GrabFood
11 Nguyen Thi Van Anh 28 collar 6 – below 12 GrabFood, Baemin
experienced
12 Bui Thi Quynh Huong 27 20 and above GrabFood, Now
employees
2. Opinions on proposed factors directly/indirectly affecting the continuance

intention to use FDAs of the Millennials in HCM city.
Throughout the course of focus group interview, almost all interviewees agree
on the direct/indirect explanative power of all proposed factors on the continuance
intention to use food delivery apps, which were specifically demonstrated in the table
below:
126
No. of interviewees agreeing

Interviewee
1 2 3 4 5 6 7 8 9 10 11 12 on explanative power of
proposed factor
Relationship
CI  PE Y Y Y Y Y Y Y Y Y Y Y Y 12
CI  EE Y Y N Y Y Y Y N Y Y Y Y 10
CI  SI Y Y Y Y Y N Y Y Y N N N 8
CI  FC Y Y Y Y N N Y Y Y Y N Y 9
CI  HM Y Y N Y Y Y Y Y Y N N N 8
CI  HB Y Y Y N Y Y Y Y N Y Y Y 10
CI  PO Y Y Y Y Y Y Y Y Y N Y Y 11
CI  SA Y Y Y Y Y Y N Y Y Y Y Y 11
SA  PE Y Y Y Y Y Y Y Y Y Y Y Y 12
SA  EE Y Y Y Y Y Y Y N Y Y Y Y 11
SA  SI Y Y Y Y Y N N Y Y N N N 7
SA  FC Y Y Y Y N N Y Y Y Y N Y 9
SA  HM Y Y N Y N Y Y N Y N N N 6
SA  PO Y Y Y Y Y Y Y Y Y N Y Y 11
SA  COF Y Y Y Y Y Y N Y Y Y Y Y 11
PE  EE Y Y Y Y Y Y Y N N Y Y Y 10
PE  TTF Y Y Y Y Y Y Y N N Y Y Y 10
PE  COF Y Y Y N Y Y Y Y Y Y Y Y 11
Among 10 factors proposed, performance expectancy (PE) was evaluated to

be have the strongest influence on users’ continuance intention and satisfaction
towards FDA usage. Regarding this, the interviewees explained the convenience and
comfort when ordering food items online and receiving them at their doorsteps
127
(typically at home and at workplace) were the main motivation behind their decision
to initially adopt and continue using FDAs on a regular basis. For some, they claimed
to enjoy the varieties of food on FDA platforms, which allowed them to try several
types of food, instead of sticking to some limited options of restaurants surrounding
their home/workplaces. Especially for office employees, they habitually order food
during lunch time, thanks to which, they can have nutritional meals while saving
additional relaxing time for lunch break.
On the other hand, hedonic motivation (HM) emerged as the least influential
determinant of FDA users’ satisfaction (SA), according to focus group interviewees.
For those who did not agree on the relationship between HM and SA, as they have
used FDAs for quite a while, they do not find these platforms novel or innovative
anymore; hence, they hardly feel excited or amused when using them. Besides, they
claimed to adopt FDAs to simply order food without engaging much into promotional
campaigns or trying to receive discount offers, from which it could be concluded that
their satisfaction was derived much from FDA performance utilities rather its hedonic
aspects.
Since all factors, together with proposed relationships between them, were
agreed on by more than half interviewees, the author decided to keep them all and
proceeded to the next stage of pilot study.
3. Opinions on measurement items underlying each factor directly/indirectly

affecting the continuance intention to use FDAs of the Millennials in HCM city.
Concerning specific measurement items for each factor, all interviewees

agreed that they were well defined, clearly understood and presented in a consistent
manner. Meanwhile, one person suggested abbreviating the phrase “(các) ứng dụng
giao thức ăn trực tuyến” into “FDA(s)” to make all measurement statements more
concise. However, many others opposed this view, saying that the abbreviation may
result in confusion when conducting the survey on a larger scale without in-person
guidance as in focus group interview. Indeed, while the abbreviation helps to shorten
all statement, chances of mix-up are high, especially when FDA has already
abbreviated for other phrases, such as the Food & Drug Administration. Furthermore,
128
the repetition of the phrase “(các) ứng dụng giao thức ăn trực tuyến” can continuously
remind respondents and underline its importance as the main convey object.
Taking all these into consideration, the author decided to keep all measurement
statements unchanged, yet would note this given point and try to gauge whether the
long phrase leaves negative any impact when respondents answer the survey or not.
129
Appendix 12: Cronbach’s Alpha coefficients for pilot study3
1. Effort Expectancy (EE)
Case Processing Summary
N %
Valid 40 100.0
Cases Excludeda 0 .0
Total 40 100.0
a. Listwise deletion based on all variables in the

procedure.
Reliability Statistics
Cronbach's N of Items
Alpha
.833 4
Item-Total Statistics
Scale Mean if Scale Variance Corrected Item- Cronbach's

Item Deleted if Item Deleted Total Alpha if Item
Correlation Deleted
EE1 13.23 4.230 .672 .793

EE2 13.30 3.549 .637 .807
EE3 13.20 3.703 .693 .775
EE4 13.45 3.741 .677 .783
2. Social Influence (SI)
N %
Valid 40 100.0
Total 40 100.0

procedure.
3
130
Alpha
.874 4

Correlation Deleted
SI1 11.30 5.600 .721 .845

SI2 11.20 4.523 .767 .828
SI3 11.15 4.849 .782 .817
SI4 11.20 5.703 .678 .859
3. Facilitating Condition (FC)
N %
Valid 40 100.0
Total 40 100.0

procedure.
Alpha
.839 4

Correlation Deleted
FC1 13.13 3.446 .643 .808

131
FC2 13.33 3.199 .764 .755

FC3 13.33 3.404 .663 .800
FC4 13.40 3.323 .622 .819
4. Hedonic Motivation (HM)
N %
Valid 40 100.0
Total 40 100.0

procedure.
Alpha
.751 3

Correlation Deleted
HM1 7.60 1.426 .656 .577

HM2 7.58 2.046 .454 .800
HM3 7.33 1.815 .658 .591
5. Price-Saving Orientation (PO)
N %
Valid 40 100.0
Total 40 100.0
132

procedure.
Alpha
.794 3

Correlation Deleted
PO1 8.25 1.577 .675 .676

PO2 8.13 1.702 .624 .732
PO3 8.08 1.661 .610 .747
6. Perceived Task-Technology Fit (TTF)
N %
Valid 40 100.0
Total 40 100.0

procedure.
Alpha
.901 3

Correlation Deleted
133
TTF1 7.50 3.026 .837 .838

TTF2 7.58 2.661 .788 .875
TTF3 7.73 2.769 .797 .863
7. Confirmation (COF)
N %
Valid 40 100.0
Total 40 100.0
procedure.
Alpha
.898 4

Correlation Deleted
COF1 11.80 4.728 .726 .888

COF2 12.00 4.821 .760 .874
COF3 11.98 4.538 .839 .844
COF4 12.08 4.994 .778 .869
134
8. Performance Expectancy (PE)
N %
Valid 40 100.0
Total 40 100.0

procedure.
Alpha
.843 4

Correlation Deleted
PE1 11.78 3.153 .701 .790

PE2 11.80 3.138 .681 .799
PE3 11.88 3.292 .642 .816
PE4 11.88 2.984 .688 .797
9. Habit (HB)
N %
Valid 40 100.0
Total 40 100.0

procedure.
135
Alpha
.918 4

Correlation Deleted
HB1 10.40 6.554 .850 .880

HB2 10.25 7.115 .813 .894
HB3 10.18 6.763 .835 .885
HB4 10.28 6.974 .753 .914
10. Satisfaction (SA)
N %
Valid 40 100.0
Total 40 100.0

procedure.
Alpha
.809 4

Correlation Deleted
SA1 11.75 6.910 .687 .729
SA2 11.58 6.558 .674 .737
SA3 11.55 7.638 .603 .771
SA4 11.40 8.092 .546 .795
136
11. Continuance Intention (CI)
N %
Valid 40 100.0
Total 40 100.0

procedure.
Alpha
.862 4

Correlation Deleted
CI1 10.65 6.028 .723 .820

CI2 10.48 5.692 .766 .800
CI3 10.55 5.690 .725 .818
CI4 10.70 6.267 .628 .856
137
Appendix 13: Adjusted measurement scales
Note: Adjustments to the original measurement scales were italicized.
Item Statement CA Source

Performance Expectancy (PE)
I feel that food delivery apps (FDAs) are useful for 0.881
PE1
ordering and receiving delivery food Bhattacherjee
(Zhao and
I feel FDAs are convenient to order and receive (2001),
PE2 Bacao,
delivery food Venkatesh et
2020)
Using FDAs improves the process of ordering and al. (2003),
PE3
receiving delivery food (Zhao and
Using FDAs improves the efficiency of ordering and Bacao, 2020)

PE4
receiving delivery food
Effort Expectancy (EE)
EE1 Learning how to use FDAs is easy 0.883 Venkatesh et
EE2 Interaction with FDAs is clear and comprehensible (Zhao and al. (2003),
EE3 It is easy to follow all the steps of FDAs Bacao, (Zhao and
EE4 It is easy to become skillful at using FDAs 2020) Bacao, 2020)
Social Influence (SI)
SI1 People who are important to me (family members, 0.860 Venkatesh et
friends, colleagues, influencers, etc.) recommend (Zhao and al. (2003),
that I use/continue using FDAs Bacao, (Zhao and
SI2 People who are important to me think FDAs are 2020) Bacao, 2020)
beneficial
SI3 People who are important to me think it is a good
idea to use/continue using FDAs
SI4 People who are important to me support me to
use/continue using FDAs
Facilitating Conditions (FC)
138
FC1 I believe that I have the necessary smartphone to use 0.83 Venkatesh et
FDAs (Palau- al. (2012),
FC2 I believe that I have the necessary knowledge to use Saumell Palau-Saumell
FDAs et al., et al. (2019)
FC3 I believe FDAs are compatible with other 2019)
technologies I use
FC4 I feel comfortable using FDAs
Hedonic Motivation (HM)
HM1 I believe that using FDAs is fun 0.916 Venkatesh et
HM2 I believe that using FDAs is enjoyable Won Lee al. (2012), Won
HM3 at al. Lee at al.
I believe that using FDAs is very entertaining (2019) (2019)
Price-Saving Orientation (PO)
PO1 I can save money by examining being informed and 0.82 Jensen (2012),
comparing the prices of different restaurants when Palau- Wen (2012),
using FDAs Saumell Palau-Saumell
PO2 I like to search for cheap restaurant deals when using et al. et al. (2019)
FDAs (2019)
PO3 I believe FDAs offer better value for my money
(something that is well worth the money spent on it)
Habit (HB)
HB1 The use of FDAs has become a habit for me. 0.889 Venkatesh et
HB2 I am in favor of using FDAs Won Lee al. (2012), Won
HB3 I feel the need to use FDAs at al. Lee at al.
HB4 Using FDAs on my smartphone has become natural (2019) (2019)
to me.
Perceived task-technology fit (TTF)
TTF1 The functions of FDAs are enough for me to order 0.880 Goodhue and
and receive the delivery food. (Zhao and Thompson
139
TTF2 The functions of FDAs are appropriate to help Bacao, (1995), (Zhao
manage the ordering and receiving the delivery of 2020) and Bacao,
food 2020)
TTF3 The functions of FDAs fully meet my requirements
of ordering and receiving the delivery of food
Confirmation (COF)
COF1 My experience with using FDAs is better than what 0.848 Bhattacherjee
I expected. (Zhao and (2001), (Zhao
COF2 The functions of FDAs are more than what I Bacao, and Bacao,
expected. 2020) 2020)
COF3 The service provided by FDAs is better than what I
expected
COF4 Overall, most of my expectations from using FDAs
were confirmed.
Satisfaction (SA)
SA1 I am very satisfied that FDAs meet my requirements 0.848 Bhattacherjee
SA2 I am satisfied with FDAs efficiency (Zhao and (2001), Hung et
SA3 My interaction with the FDAs is very satisfying. Bacao, al. (2012)
SA4 I think I did the right thing by using FDAs 2020)
CI1 I intend to continue using FDAs in the future. 0.916 Bhattacherjee
CI2 I will always try to use FDAs in my daily life (A.A. (2001),
CI3 I plan to continue to use FDAs frequently Alalwan, Venkatesh et
CI4 I have decided to use FDAs for purchasing foods the 2020) al. (2012)
next time
140
Appendix 14: Official questionnaire
BẢNG KHẢO SÁT
Ý Định Tiếp Tục Sử Dụng Các Ứng Dụng Giao Thức Ăn Trực Tuyến Của Thế
Hệ Millennial Tại Khu Vực TP.HCM
Thân gửi Quý Anh/Chị,
Tôi tên là Nguyễn Thị Thuận An, hiện đang là sinh viên năm cuối trường Đại học
Ngoại thương cơ sở II tại TP.HCM. Hiện tại tôi đang thực hiện khóa luận tốt nghiệp
với đề tài “Nghiên cứu các nhân tố ảnh hưởng đến ý định tiếp tục sử dụng các ứng
dụng giao thức ăn trực tuyến của thế hệ millennial tại khu vực TP.HCM” và rất
mong nhận được sự hợp tác của Anh/Chị trong việc hoàn thành nghiên cứu này. Tất
cả các câu trả lời trung thực của Anh/Chị sẽ giúp ích rất nhiều cho nghiên cứu.
Mỗi người tham gia sau khi hoàn thành khảo sát sẽ được tặng:
1. Tuyển tập 07 Tạp chí Chuyên ngành Marketing GAM7 (dạng Ebook);
2. Tổng hợp các báo cáo ngành/thị trường của các công ty công nghệ và công ty
nghiên cứu thị trường hàng đầu (Facebook, Deloitte, Nielsen, Kantar World
Panel, Mintel, etc);
3. Tổng hợp các tài liệu ôn thi Management Trainee, bao gồm tài liệu về Business
Case, Aptitude Test, CV Tips và một số sách chuyên ngành;
4. Tổng hợp một số template trình chiếu đẹp và chuyên nghiệp (tặng kèm fonts);
5. Đặc biệt, ba bạn may mắn nhất sẽ nhận được quyển sách GAM7 BOOK NO.6
INSIGHT - SỰ THẬT NGẦM HIỂU hoặc tài liệu khóa học BRAND
MARKETING PLAN của BrandsVietnam, hoặc tài liệu khóa học ACCOUNT
MANAGEMENT của BrandsVietnam.
Tất cả các thông tin sẽ tuyệt đối được bảo mật và chỉ nhằm mục đích nghiên cứu.
Xin chân thành cảm ơn anh/chị!
I. CÂU HỎI THÔNG TIN CHUNG

1. Anh/chị có từng sử dụng các ứng dụng giao thức ăn trực tuyến chưa?
141
⃝ Có
⃝ Không
Nếu câu trả lời của anh/chị là “Có”, vui lòng bỏ qua câu hỏi này và tiếp tục
chuyển qua các câu hỏi tiếp theo. Nếu câu trả lời của anh/chị là “Không”, vui
lòng nêu lí do vì sao anh/chị chưa sử dụng các ứng dụng này?
………………………………………………………………………………………
…..
2. Anh/chị đã sử dụng ứng dụng giao thức ăn trực tuyến trong bao lâu:
⃝ Dưới 6 tháng
⃝ Từ 6 tháng – dưới 1 năm
⃝ Từ 1 năm – dưới 2 năm
⃝ Trên 2 năm
3. Mức độ sử dụng các ứng dụng giao thức ăn trực tuyến của anh/chị?
⃝ Trên 20 lần mỗi tháng
⃝ Từ 12 đến dưới 20 lần mỗi tháng
⃝ Từ 6 đến dưới 12 lần mỗi tháng
⃝ Từ 3 đến 6 lần mỗi tháng
⃝ Dưới 3 lần mỗi tháng
Lưu ý: Số lần tương ứng với số đơn hàng đặt.
4. Anh/chị đang sử dụng (các) ứng dụng giao thức ăn trực tuyến nào? (Có
thể lựa chọn nhiều câu trả lời)
⃝ GrabFood
⃝ Now
⃝ GoFood
⃝ Baemin
⃝ Loship
⃝ Khác
5. Email để gửi quà tặng (Dùng để liên hệ để trao quà tặng): ...........................
6. Số điện thoại (Dùng để liên hệ để trao quà tặng): ..........................................
142
II. CÂU HỎI CHÍNH
Xin anh/chị vui lòng cho biết mức độ đồng ý hoặc không đồng ý cho các phát biểu
sau đây về những nhận định liên quan đến (các) ứng dụng giao thức ăn trực tuyến
bằng cách khoanh tròn vào MỘT con số ở mỗi phát biểu.
Từ trái sang phải, mức độ đồng ý của anh/chị sẽ tăng dần theo quy ước:
1: Hoàn toàn không đồng ý; 2: Không đồng ý; 3: Trung lập; 4: Đồng ý; 5: Hoàn
toàn đồng ý.
STT Phát biểu Mức độ đánh giá
Tôi thấy các ứng dụng giao thức ăn trực tuyến có

1 1 2 3 4 5
ích cho việc đặt và nhận thức ăn của tôi
Tôi thấy các ứng dụng giao thức ăn trực tuyến

2 1 2 3 4 5
tiện lợi để đặt và nhận thức ăn của tôi
Tôi thấy các ứng dụng giao thức ăn trực tuyến cải
3 1 2 3 4 5
tiến quy trình đặt và nhận thức ăn của tôi
Tôi thấy các ứng dụng giao thức ăn trực tuyến

4 1 2 3 4 5
nâng cao hiệu quả việc đặt và nhận thức ăn của tôi
Tôi thấy mình có thể dễ dàng học cách sử dụng

5 1 2 3 4 5
các ứng dụng giao thức ăn trực tuyến
Tôi thấy các tương tác khi sử dụng các ứng dụng
6 1 2 3 4 5
giao thức ăn trực tuyến rất rõ ràng và dễ hiểu
Tôi thấy các bước đặt thức ăn để giao trên các

7 ứng dụng giao thức ăn trực tuyến rất dễ để làm 1 2 3 4 5
theo
Tôi thấy dễ dàng để sử dụng các ứng dụng giao

8 1 2 3 4 5
thức ăn trực tuyến một cách thành thục
143
Những người quan trọng với tôi (gia đình, bạn bè,
9 đồng nghiệp, Influencer, v.v.) đề xuất tôi (tiếp 1 2 3 4 5
tục) sử dụng ứng dụng giao thức ăn trực tuyến
Những người quan trọng với tôi nghĩ rằng ứng

10 1 2 3 4 5
dụng giao thức ăn trực tuyến mang lợi ích cho tôi
Những người quan trọng với tôi nghĩ rằng việc

11 (tiếp tục) sử dụng ứng dụng giao thức ăn trực 1 2 3 4 5
tuyến là một ý kiến hay
Những người quan trọng với tôi ủng hộ tôi trong

12 việc (tiếp tục) sử dụng ứng dụng giao thức ăn trực 1 2 3 4 5
tuyến
Tôi có điện thoại thông minh để sử dụng ứng

13 1 2 3 4 5
dụng giao thức ăn trực tuyến
Tôi thấy tôi có đủ kiến thức để sử dụng ứng dụng

14 1 2 3 4 5
giao thức ăn trực tuyến
Tôi thấy ứng dụng giao thức ăn trực tuyến tương

15 1 2 3 4 5
thích với những công nghệ khác mà tôi sử dụng
Tôi thấy tôi hoàn toàn thoải mái khi sử dụng ứng
16 1 2 3 4 5
Tôi thấy việc sử dụng ứng dụng giao thức ăn trực

17 1 2 3 4 5
tuyến vui
Tôi thấy việc sử dụng ứng dụng giao thức ăn trực

18 1 2 3 4 5
tuyến thú vị
Tôi tận hưởng trải nghiệm sử dụng các ứng dụng

19 1 2 3 4 5
giao thức ăn trực tuyến
144
Tôi có thể tiết kiệm tiền khi được cung cấp thông
20 tin và so sánh giá giữa các cửa hàng khác nhau 1 2 3 4 5
trên ứng dụng giao thức ăn trực tuyến
Tôi thích việc tìm kiếm và sử dụng các ưu đãi khi

21 1 2 3 4 5
sử dụng ứng dụng giao thức ăn trực tuyến
Tôi thấy trong cùng một khoản chi phí, việc sử

22 dụng ứng dụng giao thức ăn trực tuyến mang đến 1 2 3 4 5
những giá trị tiêu dùng tốt hơn cho tôi
Việc sử dụng ứng dụng giao thức ăn trực tuyến đã

23 1 2 3 4 5
trở thành thói quen đối với tôi
Tôi thích sử dụng ứng dụng giao thức ăn trực

24 1 2 3 4 5
tuyến
Tôi thấy tôi cần sử dụng ứng dụng giao thức ăn

25 1 2 3 4 5
trực tuyến
Việc sử dụng ứng dụng giao thức ăn trực tuyến

26 1 2 3 4 5
trở nên rất bình thường đối với tôi
Các tính năng trên ứng dụng giao thức ăn trực

27 1 2 3 4 5
tuyến đủ để tôi đặt và nhận thức ăn

28 tuyến phù hợp để tôi kiểm tra quy trình đặt và 1 2 3 4 5
nhận thức ăn

29 tuyến đáp ứng đầy đủ các yêu cầu của tôi về đặt 1 2 3 4 5
và nhận thức ăn
Trải nghiệm sử dụng ứng dụng giao thức ăn trực

30 1 2 3 4 5
tuyến tốt hơn tôi mong đợi
145

31 1 2 3 4 5
tuyến nhiều hơn những gì tôi mong đợi
Dịch vụ được cung cấp bởi ứng dụng giao thức ăn

32 1 2 3 4 5
trực tuyến tốt hơn tôi mong đợi
Nhìn chung, hầu hết những mong đợi của tôi khi
33 sử dụng ứng dụng giao thức ăn trực tuyến được 1 2 3 4 5
đáp ứng
Tôi cảm thấy hài lòng khi các ứng dụng giao thức
34 1 2 3 4 5
ăn trực tuyến đáp ứng những yêu cầu của tôi
Tôi cảm thấy hài lòng với sự hiệu quả của các ứng
35 1 2 3 4 5
Tương tác của tôi với các ứng dụng giao thức ăn
36 1 2 3 4 5
trực tuyến làm tôi hài lòng
Tôi tin rằng quyết định sử dụng các ứng dụng

37 1 2 3 4 5
giao thức ăn trực tuyến đã là điều đúng đắn
Tôi định sẽ tiếp tục sử dụng các ứng dụng giao

38 1 2 3 4 5
thức ăn trực tuyến trong tương lai
Tôi sẽ luôn cố gắng sử dụng các ứng dụng giao

39 1 2 3 4 5
thức ăn trực tuyến trong cuộc sống hàng ngày
Tôi có kế hoạch sử dụng các ứng dụng giao thức

40 1 2 3 4 5
ăn trực tuyến thường xuyên
Tôi đã quyết định sẽ sử dụng các ứng dụng giao

41 1 2 3 4 5
thức ăn trực tuyến để đặt thức ăn cho lần tiếp theo
146
III. THÔNG TIN CÁ NHÂN

1. Giới tính của anh/chị:
⃝ Nam
⃝ Nữ
⃝ Khác
2. Tuổi của anh/chị:
⃝ Dưới 20
⃝ Từ 20 – dưới 25
⃝ Từ 25 – dưới 30
⃝ Từ 30 - dưới 35
⃝ Từ 35 - dưới 40
⃝ Từ 45 tuổi trở lên
3. Thu nhập 1 tháng:
⃝ Dưới 5.000.000 VND
⃝ Từ 5.000.000 đến dưới 10.000.000 VND
⃝ Từ 10.000.000 đến dưới 15.000.000 VND
⃝ Từ 15.000.000 đến dưới 20.000.000 VND
⃝ Từ 20.000.000 VND trở lên
4. Trình độ học vấn của anh chị:
⃝ THPT
⃝ Đại học/Cao đẳng
⃝ Thạc sĩ
⃝ Tiến sĩ
⃝ Khác.
5. Nơi sinh sống và làm việc:
⃝ TPHCM
⃝ Khác
147
KẾT THÚC KHẢO SÁT
1. Theo quan điểm của anh/chị, các ứng dụng giao thức ăn trực tuyến nên cải
thiện như thế nào để mang đến trải nghiệm tốt hơn cho người dùng?
.......................................................................................................................................
.......................................................................................................................................
2. Xin cảm ơn anh/chị đã dành thời gian hoàn thành khảo sát. Anh/chị còn
muốn chia sẻ, đóng góp, đưa ý kiến gì khác về bài khảo sát không ạ?
.......................................................................................................................................
.......................................................................................................................................
XIN CHÂN THÀNH CẢM ƠN!

148
Appendix 15: Descriptive statistics analysis by demographics factors4

Appendix 15.1: Summary of descriptive statistics analysis by demographics
factors
Valid Percent
Variable Frequency (N)
(%)
Male 145 31.7
Gender Female 312 68.3
Total 457 100.0
20 – below 25 184 40.3
25 – below 30 111 24.3
Age 30 - below 35 87 19.0
35 - 40 75 16.4
Total 457 100.0
Below 5,000,000 VND 34 7.4
5,000,000 VND - below
84 18.4
10,000,000 VND
10,000,000 VND - below
219 47.9
Monthly 15,000,000 VND
income 15,000,000 - below
90 19.7
20,000,000 VND
20,000,000 VND and
30 6.6
above
Total 457 100.0
High school 23 5.0
University/College 391 85.6
Education Master 36 7.9
background Doctor of Philosophy
7 1.5
(PhD) or Post Doc
Total 457 100.0
4
149
Appendix 15.2: Descriptive statistics analysis by demographics factors
Gender
Frequency Percent Valid Percent Cumulative

Percent
Male 145 31.7 31.7 31.7
Valid Female 312 68.3 68.3 100.0
Total 457 100.0 100.0
Age

Percent
20 – below 25 184 40.3 40.3 40.3
25 – below 30 111 24.3 24.3 64.6
Valid 30 - below 35 87 19.0 19.0 83.6
35 - 40 75 16.4 16.4 100.0
Total 457 100.0 100.0
Education

Percent
High school 23 5.0 5.0 5.0
University/College 391 85.6 85.6 90.6
Valid Master 36 7.9 7.9 98.5
Doc 7 1.5 1.5 100.0
Total 457 100.0 100.0
Income
150

Percent
Below 5,000,000 VND 34 7.4 7.4 7.4
5,000,000 VND - below

84 18.4 18.4 25.8
10,000,000 VND
10,000,000 VND - below

219 47.9 47.9 73.7
Valid 15,000,000 VND
15,000,000 - below 20,000,000

90 19.7 19.7 93.4
VND
20,000,000 VND above 30 6.6 6.6 100.0
Total 457 100.0 100.0
Location

Percent
Valid HCMC 457 100.0 100.0 100.0

151
Appendix 16: Descriptive statistics analysis by measurement scales5
Descriptive Statistics
N Minimum Maximum Mean Std. Deviation
EE1 457 2 5 4.35 .798

EE2 457 1 5 4.39 .799
EE3 457 2 5 4.41 .796
EE4 457 1 5 4.37 .795
SI1 457 1 5 3.65 .775
SI2 457 1 5 3.67 .826
SI3 457 1 5 3.69 .825
SI4 457 1 5 3.69 .784
FC1 457 1 5 4.42 .754
FC2 457 1 5 4.41 .765
FC3 457 1 5 4.40 .798
FC4 457 1 5 4.45 .762
HM1 457 1 5 3.62 .808
HM2 457 1 5 3.59 .779
HM3 457 1 5 3.63 .793
PO1 457 1 5 4.05 .768
PO2 457 1 5 4.05 .768
PO3 457 1 5 4.08 .777
TTF1 457 2 5 4.21 .703
TTF2 457 1 5 4.21 .702
TTF3 457 1 5 4.22 .747
COF1 457 2 5 4.14 .737
COF2 457 1 5 4.10 .748
COF3 457 1 5 4.13 .705
COF4 457 2 5 4.12 .727
PE1 457 1 5 4.33 .736
PE2 457 2 5 4.34 .768
PE3 457 1 5 4.32 .774
PE4 457 1 5 4.36 .772
HB1 457 1 5 4.04 .813
HB2 457 1 5 4.02 .775
HB3 457 1 5 4.07 .792
HB4 457 1 5 4.04 .776
SA1 457 1 5 4.23 .870
SA2 457 1 5 4.21 .889
SA3 457 1 5 4.26 .851
SA4 457 1 5 4.25 .840
5
152
CI1 457 1 5 4.06 .721

CI2 457 1 5 4.10 .767
CI3 457 1 5 4.13 .771
CI4 457 1 5 4.11 .792
Valid N (listwise) 457
153
Appendix 17: Cronbach’s Alpha coefficients for official study6
Appendix 17.1: Summary of Cronbach’s Alpha coefficients for official study

Corrected Cronbach's
Cronbach's Measurement
Construct Item-Total Alpha if Item
Alpha items
Correlation Deleted
.827 PE1 .627 .794
Performance
PE2 .629 .793
Expectancy
PE3 .686 .767
(PE)
PE4 .671 .774
.800 EE1 .637 .738
Effort
EE2 .605 .754
Expectancy
EE3 .594 .760
(EE)
EE4 .616 .749
.855 SI1 .697 .816
Social SI2 .683 .822
Influence (SI) SI3 .727 .803
SI4 .684 .821
.805 FC1 .621 .755
Facilitating FC2 .598 .766
Condition (FC) FC3 .660 .736
FC4 .601 .765
Hedonic .806 HM1 .661 .726
Motivation HM2 .645 .743
(HM) HM3 .653 .734
Price-Saving .816 PO1 .678 .736
Orientation PO2 .644 .770
(PO) PO3 .679 .734
Habit (HB) .819 HB1 .640 .772
6
154
HB2 .655 .765

HB3 .665 .760
HB4 .601 .790
Perceived .794 TTF1 .618 .740
Task- TTF2 .623 .735
Technology Fit .671 .683
TTF3
(TTF)
.841 COF1 .682 .795
Confirmation COF2 .674 .798
(COF) COF3 .667 .801
COF4 .673 .799
.833 SA1 .682 .779
Satisfaction SA2 .647 .796
(SA) SA3 .655 .791
SA4 .663 .788
.873 CI1 .721 .841
Continuance
CI2 .727 .838
Intention (CI)
CI3 .745 .831
CI4 .721 .841
155
Appendix 17.2: Cronbach’s Alpha coefficients for official study
1. Effort Expectancy (EE)
N %
Valid 457 100.0
Total 457 100.0

procedure.
Alpha
.800 4

Correlation Deleted
EE1 13.18 3.750 .637 .738

EE2 13.13 3.826 .605 .754
EE3 13.11 3.864 .594 .760
EE4 13.16 3.810 .616 .749
2. Social Influence (SI)
N %
Valid 457 100.0
Total 457 100.0

procedure.
156
Alpha
.855 4

Correlation Deleted
SI1 11.04 4.336 .697 .816

SI2 11.03 4.192 .683 .822
SI3 11.01 4.083 .727 .803
SI4 11.01 4.338 .684 .821
3. Facilitating Condition (FC)
N %
Valid 457 100.0
Total 457 100.0

procedure.
Alpha
.805 4

Correlation Deleted
157
FC1 13.26 3.629 .621 .755

FC2 13.27 3.651 .598 .766
FC3 13.28 3.405 .660 .736
FC4 13.24 3.652 .601 .765
4. Hedonic Motivation (HM)
N %
Valid 457 100.0
Total 457 100.0

procedure.
Alpha
.806 3

Correlation Deleted
HM1 7.22 1.931 .665 .721

HM2 7.25 2.047 .639 .748
HM3 7.21 1.986 .655 .732
5. Price-Saving Orientation (PO)
N %
Valid 457 100.0
Cases
Excludeda 0 .0
158
Total 457 100.0

procedure.
Alpha
.816 3

Correlation Deleted
PO1 8.13 1.887 .678 .736

PO2 8.13 1.940 .644 .770
PO3 8.10 1.864 .679 .734
6. Perceived Task-Technology Fit (TTF)
N %
Valid 457 100.0
Total 457 100.0

procedure.
Alpha
.794 3
159

Correlation Deleted
TTF1 8.43 1.667 .618 .740

TTF2 8.43 1.662 .623 .735
TTF3 8.42 1.499 .671 .683
7. Confirmation (COF)
N %
Valid 457 100.0
Total 457 100.0

procedure.
Alpha
.841 4

Correlation Deleted
COF1 12.34 3.371 .682 .795

COF2 12.39 3.351 .674 .798
COF3 12.36 3.503 .667 .801
COF4 12.37 3.422 .673 .799
160
8. Performance Expectancy (PE)
N %
Valid 457 100.0
Total 457 100.0

procedure.
Alpha
.827 4

Correlation Deleted
PE1 13.02 3.789 .627 .794

PE2 13.01 3.684 .629 .793
PE3 13.04 3.534 .686 .767
PE4 12.99 3.574 .671 .774
9. Habit (HB)
N %
Valid 457 100.0
Total 457 100.0

procedure.
161
Alpha
.819 4

Correlation Deleted
HB1 12.13 3.771 .640 .772

HB2 12.15 3.857 .655 .765
HB3 12.10 3.777 .665 .760
HB4 12.14 3.986 .601 .790
10. Satisfaction (SA)
N %
Valid 457 100.0
Total 457 100.0

procedure.
Alpha
.833 4

Correlation Deleted
SA1 12.72 4.621 .682 .779

SA2 12.74 4.660 .647 .796
SA3 12.69 4.770 .655 .791
SA4 12.71 4.791 .663 .788
162
11. Continuance Intention (CI)
N %
Valid 457 100.0
Total 457 100.0

procedure.
Alpha
.873 4

Correlation Deleted
CI1 12.34 4.119 .721 .841

CI2 12.30 3.945 .727 .838
CI3 12.27 3.890 .745 .831
CI4 12.30 3.875 .721 .841
163
Appendix 18: Explanatory Factor Analysis (EFA)7
Appendix 18.1: Explanatory Factor Analysis (EFA) for 10 variables affecting

KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .882

Approx. Chi-Square 7193.219
Bartlett's Test of Sphericity df 666
Sig. .000
Communalities
Initial Extraction
EE1 .482 .561

EE2 .426 .502
EE3 .397 .464
EE4 .456 .532
SI1 .542 .599
SI2 .517 .579
SI3 .588 .678
SI4 .516 .578
FC1 .455 .533
FC2 .432 .500
FC3 .516 .587
FC4 .428 .495
HM1 .504 .617
HM2 .480 .561
HM3 .479 .595
PO1 .507 .608
PO2 .478 .563
PO3 .521 .641
TTF1 .454 .526
TTF2 .430 .543
TTF3 .506 .661
COF1 .504 .615
COF2 .488 .568
COF3 .502 .570
COF4 .485 .566
7
164
PE1 .439 .483

PE2 .463 .523
PE3 .548 .616
PE4 .501 .609
HB1 .478 .545
HB2 .499 .569
HB3 .492 .594
HB4 .410 .466
SA1 .517 .612
SA2 .491 .546
SA3 .479 .542
SA4 .527 .588
Extraction Method: Principal Axis

Factoring.
Total Variance Explained
Factor Initial Eigenvalues Extraction Sums of Squared Loadings Rotation

Sums of
Squared
Loadingsa
Total % of Variance Cumulative % Total % of Variance Cumulative % Total
1 8.312 22.465 22.465 7.879 21.294 21.294 4.591

2 3.610 9.756 32.222 3.180 8.594 29.888 2.703
3 2.522 6.817 39.038 2.085 5.635 35.524 2.705
4 2.192 5.923 44.961 1.775 4.796 40.320 4.961
5 1.870 5.055 50.016 1.453 3.928 44.248 4.854
6 1.742 4.709 54.725 1.312 3.546 47.794 4.781
7 1.338 3.615 58.341 .903 2.440 50.234 3.817
8 1.302 3.519 61.859 .853 2.306 52.540 5.684
9 1.254 3.388 65.247 .830 2.243 54.782 3.831
10 1.099 2.969 68.216 .666 1.801 56.583 2.141
11 .678 1.833 70.050
12 .636 1.718 71.768
13 .606 1.639 73.406
14 .581 1.571 74.977
15 .571 1.544 76.520
16 .542 1.466 77.986
17 .532 1.439 79.425
18 .518 1.400 80.825
165
19 .485 1.311 82.136

20 .463 1.253 83.388
21 .454 1.227 84.615
22 .442 1.195 85.810
23 .438 1.184 86.994
24 .430 1.163 88.157
25 .415 1.123 89.280
26 .410 1.108 90.388
27 .387 1.047 91.435
28 .382 1.031 92.466
29 .359 .970 93.436
30 .350 .945 94.381
31 .343 .928 95.309
32 .329 .890 96.199
33 .314 .850 97.049
34 .302 .816 97.865
35 .280 .757 98.622
36 .263 .710 99.332
37 .247 .668 100.000
Extraction Method: Principal Axis Factoring.

a. When factors are correlated, sums of squared loadings cannot be added to obtain a total variance.
166
Factor Matrixa
Factor
1 2 3 4 5 6 7 8 9 10
SA4 .640
SA2 .639
PE3 .633
SA1 .627
SA3 .602
EE1 .577
FC3 .553
SI3 .550
PE4 .549
PE2 .542
PE1 .539
SI4 .539
EE4 .538
SI2 .538
SI1 .536
FC1 .522
EE3 .512
HM2 .512
EE2 .507
FC2 .503
FC4 .502
PO1
PO3
HM1
PO2
HM3
COF2 .571
COF4 .566
COF3 .555
COF1 .541
TTF3
TTF1
TTF2
HB3 .579
HB2 .540
HB1 .511
HB4

a. 10 factors extracted. 9 iterations required.
167
Pattern Matrixa
Factor
1 2 3 4 5 6 7 8 9 10
SI3 .836
SI1 .755
SI4 .740
SI2 .731
COF1 .824
COF4 .741
COF2 .733
COF3 .713
HB3 .768
HB1 .726
HB2 .714
HB4 .693
PE4 .805
PE3 .717
PE2 .685
PE1 .653
EE2 .724
EE1 .711
EE4 .698
EE3 .670
FC1 .740
FC3 .726
FC2 .692
FC4 .646
PO3 .778
PO1 .761
PO2 .752
SA1 .820
SA3 .694
SA4 .691
SA2 .641
HM3 .794
HM1 .776
HM2 .697
TTF3 .798
TTF2 .735
TTF1 .709

Rotation Method: Promax with Kaiser Normalization.
a. Rotation converged in 7 iterations.
168
Structure Matrix
Factor
1 2 3 4 5 6 7 8 9 10
SI3 .819
SI1 .770
SI2 .755
SI4 .752
COF
.776
1
COF
.752
2
COF
.749
4
COF
.748
3
HB3 .765
HB2 .748
HB1 .731
HB4 .678
PE3 .779
PE4 .773
PE2 .716
PE1 .689
EE1 .741
EE4 .718
EE2 .703
EE3 .677
FC3 .759
FC1 .725
FC4 .691
FC2 .688
PO3 .796
PO1 .778
PO2 .745
SA1 .777
SA4 .754
SA3 .732
SA2 .725
HM1 .782
HM3 .767
HM2 .742
TTF3 .809
169
TTF2 .730
TTF1 .719

Factor Correlation Matrix
Factor 1 2 3 4 5 6 7 8 9 10
1 1.000 .026 .022 .406 .387 .500 .320 .459 .351 -.011
2 .026 1.000 .102 .197 .020 -.144 .042 .243 -.011 .217
3 .022 .102 1.000 .159 .080 .078 .069 .367 .085 .212
4 .406 .197 .159 1.000 .507 .456 .331 .527 .318 .224
5 .387 .020 .080 .507 1.000 .514 .401 .531 .389 -.011
6 .500 -.144 .078 .456 .514 1.000 .320 .461 .367 .012
7 .320 .042 .069 .331 .401 .320 1.000 .440 .500 -.040
8 .459 .243 .367 .527 .531 .461 .440 1.000 .427 .169
9 .351 -.011 .085 .318 .389 .367 .500 .427 1.000 -.018
10 -.011 .217 .212 .224 -.011 .012 -.040 .169 -.018 1.000

170
Appendix 18.2: Explanatory Factor Analysis (EFA) for Continuance Intention

(CI)
KMO and Bartlett's Test
Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .835

Approx. Chi-Square 885.562
Bartlett's Test of Sphericity df 6
Sig. .000
Communalities
Initial Extraction
CI1 .520 .617

CI2 .531 .630
CI3 .557 .667
CI4 .521 .619
Extraction Method: Principal Axis

Factoring.
Total Variance Explained
Factor Initial Eigenvalues Extraction Sums of Squared Loadings
Total % of Variance Cumulative % Total % of Variance Cumulative %
1 2.900 72.501 72.501 2.534 63.340 63.340

2 .393 9.821 82.321
3 .370 9.262 91.584
4 .337 8.416 100.000
Factor Matrixa
171
Factor
CI3 .817
CI2 .794
CI4 .787
CI1 .786
Extraction Method:
Principal Axis
Factoring.
a. 1 factors
extracted. 5
iterations required.
172
Appendix 19: Confirmatory Factor Analysis (CFA)8
Appendix 19.1: CFA path diagram
Appendix 19.2: Model-fit indices of measurement model
8
173
CMIN
Model NPAR CMIN DF P CMIN/DF
Default model 136 763.600 725 .155 1.053
Saturated model 861 .000 0
Independence model 41 8878.701 820 .000 10.828
RMR, GFI
Model RMR GFI AGFI PGFI
Default model .018 .927 .913 .780
Saturated model .000 1.000
Independence model .156 .273 .237 .260
Baseline Comparisons
NFI RFI IFI TLI

Model CFI
Delta1 rho1 Delta2 rho2
Default model .914 .903 .995 .995 .995
Saturated model 1.000 1.000 1.000
Independence model .000 .000 .000 .000 .000
Parsimony-Adjusted Measures
Model PRATIO PNFI PCFI
Default model .884 .808 .880

174
Saturated model .000 .000 .000
Independence model 1.000 .000 .000
NCP
Model NCP LO 90 HI 90
Default model 38.600 .000 108.721
Independence model 8058.701 7759.216 8364.671
FMIN
Model FMIN F0 LO 90 HI 90
Default model 1.675 .085 .000 .238
Saturated model .000 .000 .000 .000
Independence model 19.471 17.673 17.016 18.344
RMSEA
Model RMSEA LO 90 HI 90 PCLOSE
Default model .011 .000 .018 1.000

175
AIC
Model AIC BCC BIC CAIC
Default model 1035.600 1063.195 1596.557 1732.557
Saturated model 1722.000 1896.696 5273.352 6134.352
ECVI
Model ECVI LO 90 HI 90 MECVI
Default model 2.271 2.186 2.425 2.332
Saturated model 3.776 3.776 3.776 4.159
HOELTER
HOELTER HOELTER
Model
.05 .01
Default model 472 488
Independence model 46 48
Minimization: .035
Miscellaneous: 1.239
Bootstrap: .000
Total: 1.274
176
Appendix 19.3: Standardized Regression Weights
Estimate
SI4 <--- SI .746
SI3 <--- SI .809
SI2 <--- SI .752
SI1 <--- SI .782
CI4 <--- CI .778
CI3 <--- CI .816
CI2 <--- CI .800
CI1 <--- CI .790
HB4 <--- HB .676
HB3 <--- HB .748
HB2 <--- HB .758
HB1 <--- HB .733
COF4 <--- COF .755
COF3 <--- COF .753
COF2 <--- COF .759
COF1 <--- COF .750
SA4 <--- SA .752
SA3 <--- SA .733
SA2 <--- SA .738

177
Estimate
SA1 <--- SA .759
EE4 <--- EE .719
EE3 <--- EE .683
EE2 <--- EE .688
EE1 <--- EE .740
FC4 <--- FC .692
FC3 <--- FC .776
FC2 <--- FC .674
FC1 <--- FC .707
PE4 <--- PE .749
PE3 <--- PE .793
PE2 <--- PE .705
PE1 <--- PE .706
PO3 <--- PO .794
PO2 <--- PO .732
PO1 <--- PO .790
HM3 <--- HM .746
HM2 <--- HM .760
HM1 <--- HM .779
TTF3 <--- TTF .819

178
Estimate
TTF2 <--- TTF .718
TTF1 <--- TTF .714

179
Appendix 19.4: Validity and reliability test
C AV MS H TT
SI CI HB COF SA EE FC PE PO
R E V M F
0.85
SI 0.597 0.294 0.773
6
0.87 0.542*
CI 0.634 0.539 0.796
4 **
0.81 0.359*
HB 0.532 0.159 0.023 0.729
9 **
CO 0.84 0.234*
0.569 0.062 0.028 0.100† 0.754
F 1 **
0.83 0.484* 0.734* 0.399* 0.249*

SA 0.556 0.539 0.745
3 ** ** ** **
0.80 0.413* 0.646* 0.561*

EE 0.501 0.417 0.102† 0.021 0.708
0 ** ** **
-
0.80 0.519* 0.589* 0.499* 0.542*
FC 0.509 0.347 0.078 0.153* 0.713
5 ** ** ** **
*
0.82 0.434* 0.682* 0.170* 0.207* 0.578* 0.547* 0.501*

PE 0.546 0.465 0.739
8 ** ** * ** ** ** **
0.81 0.335* 0.516* 0.475* 0.420* 0.326* 0.361*

PO 0.597 0.274 0.080 0.040 0.773
6 ** ** ** ** ** **
H 0.80 0.368* 0.510* 0.448* 0.404* 0.377* 0.328* 0.524*

0.580 0.274 0.086 -0.014 0.762
M 6 ** ** ** ** ** ** **
TT 0.79 0.209* 0.240* 0.239* 0.169* 0.253* -

0.565 0.064 -0.006 ⁂ 0.026 -0.031 0.752
F 5 ** ** ** * ** 0.011
180
Appendix 20: Structural Equation Modeling (SEM)9
Appendix 20.1: SEM path diagram
9
181
Appendix 20.2: Indices for model fit, validity & reliability
I. Model Fit Summary
CMIN
Model NPAR CMIN DF P CMIN/DF
Default model 128 867.775 733 .000 1.184
Saturated model 861 .000 0
Independence model 41 8878.701 820 .000 10.828
RMR, GFI
Model RMR GFI AGFI PGFI
Default model .027 .918 .903 .781
Saturated model .000 1.000
Baseline Comparisons
NFI RFI IFI TLI

Model CFI
Delta1 rho1 Delta2 rho2
Default model .902 .891 .983 .981 .983
Saturated model 1.000 1.000 1.000
Independence model .000 .000 .000 .000 .000
Parsimony-Adjusted Measures
182
Default model .894 .807 .879
Independence model 1.000 .000 .000
NCP
Model NCP LO 90 HI 90
Default model 134.775 65.218 212.570
Independence model 8058.701 7759.216 8364.671
FMIN
Model FMIN F0 LO 90 HI 90
Default model 1.903 .296 .143 .466
Saturated model .000 .000 .000 .000
RMSEA
Model RMSEA LO 90 HI 90 PCLOSE
Default model .020 .014 .025 1.000

183
AIC
Model AIC BCC BIC CAIC
Default model 1123.775 1149.746 1651.735 1779.735
Saturated model 1722.000 1896.696 5273.352 6134.352
ECVI
Model ECVI LO 90 HI 90 MECVI
Default model 2.464 2.312 2.635 2.521
Saturated model 3.776 3.776 3.776 4.159
HOELTER
HOELTER HOELTER
Model
.05 .01
Default model 419 434
Independence model 46 48
Minimization: .034
Miscellaneous: 1.406
Bootstrap: .000
Total: 1.440
184
II. Standardized Regression Weights: (Group number 1 - Default model)
Estimate
PE <--- TTF .233
PE <--- COF .142
PE <--- EE .595
SA <--- COF .238
SA <--- EE .183
SA <--- SI .123
SA <--- FC .186
SA <--- HM .120
SA <--- PE .197
SA <--- PO .158
CI <--- PE .270
CI <--- SA .234
CI <--- EE .149
CI <--- SI .135
CI <--- FC .107
CI <--- HM .104
CI <--- PO .108
CI <--- HB .197
SI4 <--- SI .745
SI3 <--- SI .811

185
Estimate
SI2 <--- SI .752
SI1 <--- SI .782
CI4 <--- CI .772
CI3 <--- CI .808
CI2 <--- CI .796
CI1 <--- CI .780
HB4 <--- HB .680
HB3 <--- HB .748
HB2 <--- HB .753
HB1 <--- HB .732
COF4 <--- COF .753
COF3 <--- COF .752
COF2 <--- COF .759
COF1 <--- COF .752
SA4 <--- SA .739
SA3 <--- SA .728
SA2 <--- SA .742
SA1 <--- SA .756
EE4 <--- EE .714
EE3 <--- EE .672

186
Estimate
EE2 <--- EE .679
EE1 <--- EE .732
FC4 <--- FC .693
FC3 <--- FC .775
FC2 <--- FC .669
FC1 <--- FC .713
PE4 <--- PE .748
PE3 <--- PE .788
PE2 <--- PE .708
PE1 <--- PE .704
PO3 <--- PO .791
PO2 <--- PO .733
PO1 <--- PO .793
HM3 <--- HM .746
HM2 <--- HM .760
HM1 <--- HM .778
TTF3 <--- TTF .817
TTF2 <--- TTF .718
TTF1 <--- TTF .715

187
Appendix 20.3: Regression Weights

Estimate S.E. C.R. P Label
PE <--- TTF .221 .050 4.455 ***
PE <--- COF .150 .053 2.841 .004
PE <--- EE .605 .062 9.789 ***
SA <--- COF .268 .055 4.872 ***
SA <--- EE .199 .087 2.282 .023
SA <--- SI .130 .059 2.201 .028
SA <--- FC .217 .079 2.762 .006
SA <--- HM .126 .061 2.054 .040
SA <--- PE .211 .067 3.152 .002
SA <--- PO .159 .059 2.716 .007
CI <--- PE .282 .055 5.091 ***
CI <--- SA .228 .055 4.151 ***
CI <--- EE .158 .070 2.274 .023
CI <--- SI .139 .047 2.934 .003
CI <--- FC .122 .060 2.026 .043
CI <--- HM .106 .049 2.172 .030
CI <--- PO .106 .047 2.250 .024
CI <--- HB .225 .043 5.166 ***

188
Appendix 20.4: Standardized Regression Weights

Estimate
PE <--- TTF .233
PE <--- COF .142
PE <--- EE .595
SA <--- COF .238
SA <--- EE .183
SA <--- SI .123
SA <--- FC .186
SA <--- HM .120
SA <--- PE .197
SA <--- PO .158
CI <--- PE .270
CI <--- SA .234
CI <--- EE .149
CI <--- SI .135
CI <--- FC .107
CI <--- HM .104
CI <--- PO .108
CI <--- HB .197
SI4 <--- SI .745
SI3 <--- SI .811

189
Estimate
SI2 <--- SI .752
SI1 <--- SI .782
CI4 <--- CI .772
CI3 <--- CI .808
CI2 <--- CI .796
CI1 <--- CI .780
HB4 <--- HB .680
HB3 <--- HB .748
HB2 <--- HB .753
HB1 <--- HB .732
COF4 <--- COF .753
COF3 <--- COF .752
COF2 <--- COF .759
COF1 <--- COF .752
SA4 <--- SA .739
SA3 <--- SA .728
SA2 <--- SA .742
SA1 <--- SA .756
EE4 <--- EE .714
EE3 <--- EE .672

190
Estimate
EE2 <--- EE .679
EE1 <--- EE .732
FC4 <--- FC .693
FC3 <--- FC .775
FC2 <--- FC .669
FC1 <--- FC .713
PE4 <--- PE .748
PE3 <--- PE .788
PE2 <--- PE .708
PE1 <--- PE .704
PO3 <--- PO .791
PO2 <--- PO .733
PO1 <--- PO .793
HM3 <--- HM .746
HM2 <--- HM .760
HM1 <--- HM .778
TTF3 <--- TTF .817
TTF2 <--- TTF .718
TTF1 <--- TTF .715

191
Appendix 20.5: Squared Multiple Correlations
Estimate
PE .448
SA .550
CI .738
TTF1 .511
TTF2 .516
TTF3 .668
HM1 .606
HM2 .578
HM3 .557
PO1 .628
PO2 .537
PO3 .625
PE1 .496
PE2 .501
PE3 .621
PE4 .560
FC1 .508
FC2 .448
FC3 .600
192
Estimate
FC4 .480
EE1 .535
EE2 .462
EE3 .452
EE4 .510
SA1 .571
SA2 .550
SA3 .529
SA4 .546
COF1 .565
COF2 .576
COF3 .565
COF4 .567
HB1 .536
HB2 .566
HB3 .560
HB4 .463
CI1 .609
CI2 .633
CI3 .654
193
Estimate
CI4 .596
SI1 .612
SI2 .565
SI3 .657
SI4 .555
194
Appendix 21: Independent Sample t-Test & One-Way ANOVA on

demographic variables10
Appendix 21.1: Gender
Independent Samples Test
Levene's Test for t-test for Equality of Means

Equality of
Variances
F Sig. t df Sig. (2- Mean Std. 95% Confidence

tailed) Differenc Error Interval of the
e Differenc Difference
e Lower Upper
Equal variances
1.996 .158 -.285 455 .776 -.01864 .06535 -.14706 .10978
assumed
CI
Equal variances 249.2
-.271 .786 -.01864 .06870 -.15394 .11666
not assumed 87
Group Statistics
Gender N Mean Std. Deviation Std. Error Mean
Male 145 4.0879 .71110 .05905

CI
Female 312 4.1066 .61992 .03510
10
195
Appendix 21.2: Age
Test of Homogeneity of Variances

CI
Levene Statistic df1 df2 Sig.
.904 3 453 .439
ANOVA
CI
Sum of Squares df Mean Square F Sig.
Between Groups 2.270 3 .757 1.803 .146

Within Groups 190.100 453 .420
Total 192.370 456
196
Appendix 21.3: Income

CI
.528 4 452 .715
ANOVA
CI
Between Groups 2.433 4 .608 1.447 .217

Total 192.370 456
197
Appendix 21.4: Educational background

CI
.089 3 453 .966
ANOVA
CI
Between Groups 1.753 3 .584 1.388 .246

Total 192.370 456

NGUYỄN THỊ THUẬN AN-1701015008 -KTĐN

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NGUYỄN THỊ THUẬN AN-1701015008 -KTĐN

Uploaded by

Copyright:

Available Formats

FOREIGN TRADE UNIVERSITY

HO CHI MINH CITY CAMPUS

Major: International Business Economics

FACTORS AFFECTING THE CONTINUANCE

Student: Nguyễn Thị Thuận An

Ho Chi Minh City, December 18th, 2020

Ho Chi Minh City, ….../….../...............

Student’s full name: NGUYỄN THỊ THUẬN AN Student code:

Ho Chi Minh City, December 2020

Nguyễn Thị Thuận An

Ho Chi Minh City, December 2020

Nguyễn Thị Thuận An

LIST OF TABLES AND FIGURES

CHAPTER 1: INTRODUCTION ............................................................................ 1

1.1. Rationale of the research ............................................................................... 1

1.2. Research aims and objectives........................................................................ 3

1.2.1. Research aims ...........................................................................................3

1.2.2. Research objectives ...................................................................................4

1.3. Subject and scope of the research ................................................................. 4

1.3.1. Research subject ........................................................................................4

1.3.2. Research scope ..........................................................................................5

1.4. Research background .................................................................................... 5

1.4.1. Foreign studies ..........................................................................................5

1.4.2. Domestic studies .......................................................................................6

1.5. Research questions ............................................................................................. 6

1.6. Research methodology ....................................................................................... 7

1.7. Contribution and significance of the research ................................................ 8

1.8. Research structure ............................................................................................. 9

SUMMARY OF CHAPTER 1 ............................................................................. 9

CHAPTER 2: THEORETICAL FRAMEWORK AND LITERATURE

2.1. Food delivery ................................................................................................ 10

2.1.1. The concept of food delivery ..................................................................10

2.1.2. Food delivery market in Vietnam ...........................................................11

2.2. Food delivery app (FDA) ............................................................................. 13

2.2.1. Food delivery app....................................................................................13

2.2.3. Food delivery app market in Vietnam .................................................... 14

2.2.4. Food delivery app usage in Vietnam ...................................................... 17

2.3. The Millennials .............................................................................................18

2.3.1. Definitions of the Millennials ................................................................ 18

2.3.2. Some key characteristics of the Millennials .......................................... 19

2.4. The concept of behavioral continuance intention to use ..........................21

2.4.1. Behavioral intention to use..................................................................... 21

2.4.2. Behavioral continuance intention to use (CI) ........................................ 22

2.5. Conceptual models and theories on factors affecting continuance

2.5.1. Theory of Reasoned Action (TRA) (1975) ............................................ 24

2.5.2. Technology Acceptance Model (TAM) (1989) ..................................... 24

2.5.3. Task-Technology Fit (TTF) Theory (1995) ........................................... 25

2.5.4. Unified Theory of Acceptance and Use of Technology (UTAUT) (2003)

2.5.5. Unified Theory of Acceptance and Use of Technology 2 (UTAUT2)

2.5.6. Expectancy Confirmation Model (ECM) (2001) ................................... 27

2.6. Empirical research on continuance intention to use food delivery apps 28

2.6.1. Foreign studies ....................................................................................... 28

2.6.2. Domestic studies .................................................................................... 30

2.7. Hypothesis development and proposed research model ...........................30

2.7.1. Rationale for the proposed research model ............................................ 30

2.7.2. The proposed research model................................................................. 32

2.7.3. Hypothesis development ........................................................................ 32

CHAPTER 3: RESEARCH METHODOLOGY ................................................. 42

3.1. Research process .......................................................................................... 42

3.2. Preliminary research ................................................................................... 43

3.2.1. Initial questionnaire development ...........................................................43

3.2.2. Pretest: Expert reviews and focus groups ...............................................45

3.2.3. Pilot study................................................................................................47