ABSTRACT

In this modern world, technology has evolved into most of our daily activities. Because of the
advancement in living standards, people nowadays are inclined to buy more to satisfy their increasing
needs. Seeing this opportunity, businessmen have built up a model called “Buy Now, Pay Later” to
motivate consumers to spend more with fewer extra fees. For example, the “Shopee Buy Now, Pay
Later” feature, one of four payment options offered by Shopee application, is a payment solution
allowing buyers to make immediate purchases with no interest within specific periods of time.
However, it is still controversial whether the feature “Buy Now, Pay Later” can make people spend
money impulsively on unnecessary purchases. Therefore,the researchers carried out an investigation on
the subject “Factors that contribute to the impulsive buying tendency of students using BNPL services”
to measure which elements of the feature contribute to people’s impulsive buying tendency and suggest
some remedies for them to improve their financial conditions.

The topic is based on the analysis of factors that influence buying intention of students when
using Buy Now, Pay Later services, including Customization design, Perceived Security, Simple and
convenient usage, Awareness of Buy Now, Pay Later, User friendliness of Buy Now, Pay Later. With
the data gathered from an online survey conducted by a sample of 175 students in UEH university, the
results showed that Awareness of Buy Now, Pay Later, Perceived Security, User friendliness of Buy
Now, Pay Later have a significant effect on overspending tendency of purchasers, of which the
Awareness has the most remarkable influence. However, remaining factors like Simple and convenient
usage and User friendliness of Buy Now, Pay Later barely contribute to buyers' impulsive purchasing
behaviors. Relying on the outcome obtained, the authors have suggested some measures to help UEH
students to create a realistic budget and allocate funds efficiently in order to achieve financial stability
in future.

1
 TABLE OF CONTENTS
ABSTRACT...........................................................................................................
TABLE OF CONTENTS...........................................................................................
TABLES LIST........................................................................................................
FIGURES LIST......................................................................................................
ABBREVIATIONS LIST...........................................................................................
CHAPTER I: INTRODUCTION..................................................................................
1.1. Reason (Problem Statement).........................................................................................
1.2. Aims of research............................................................................................................
1.3. Research questions.......................................................................................................
1.4. Research methods.........................................................................................................
CHAPTER II: THEORETICAL BASIS.........................................................................
2.1. Related research...........................................................................................................
2.2. Literature review...........................................................................................................
2.2.1. Buy Now, Pay Later................................................................................................
2.2.2. Impulsive buying tendency....................................................................................
2.2.3. Interface Design.....................................................................................................
2.2.4. Perceived Control...................................................................................................
2.2.5. Perceived Security.................................................................................................
2.3. Research Hypotheses....................................................................................................
2.3.1. Dependent variable: Impulsive Buying Tendency..................................................
2.3.2. The impact of BNPL’s Customization Design on users’ impulsive
buying tendency..............................................................................................................
2.3.3. The impact of BNPL’s Perceived Security on users’ impulsive buying
tendency..........................................................................................................................
2.3.4. The impact of BNPL’s Simple and Convenient usage on users’
impulsive buying tendency..............................................................................................
2.3.5. The impact of BNPL’s Awareness on users’ impulsive buying tendency
.........................................................................................................................................
2.3.6. The impact of BNPL’s User-Friendliness on users’ impulsive buying
tendency..........................................................................................................................
2.4. Proposed Research Model.............................................................................................
CHAPTER III: METHODOLOGY..............................................................................
3.1. Research process..........................................................................................................
The overview of main steps in the research procedure is illustrated in Figure 2.
.............................................................................................................................................
3.2. Research subject and scope..........................................................................................
3.2.1. Research sample....................................................................................................
3.2.2. Research scope......................................................................................................
3.3. Research methods.........................................................................................................

2
 3.3.1. Qualitative research...............................................................................................
3.3.2. Quantitative research............................................................................................
3.4. Sample size...................................................................................................................
3.5. Sampling technique.......................................................................................................
3.6. Procedure of collecting data..........................................................................................
3.7. Measurement Scales.....................................................................................................
3.7.1. “Customization design” scale................................................................................
3.7.2. “Perceived Security” scale.....................................................................................
3.7.3. “Simple and convenient usage” scale....................................................................
3.7.4. “Awareness of BNPL” scale....................................................................................
3.7.5. “User friendliness of BNPL” scale...........................................................................
3.7.6. “Impulsive Buying Tendency” scale.......................................................................
3.8.1. Descriptive Statistics..............................................................................................
3.8.2. Measure Reliability Assessment.............................................................................
3.8.3. Exploratory Factor Analysis....................................................................................
3.8.4. Pearson Correlation Analysis..................................................................................
3.8.5. Multivariate Regression Analysis............................................................................
CHAPTER IV: RESEARCH RESULTS.......................................................................
4.1. Descriptive statistics of the survey sample...................................................................
4.2. Descriptive statistics of the independent variables.......................................................
4.3. Cronbach’s Alpha: Scale Reliability Test........................................................................
4.4. Exploratory Factor Analysis: Scale Validity Test............................................................
4.5. Pearson Correlation: Testing research hypotheses.......................................................
4.6. Multivariate Regression Analysis...................................................................................
4.6.1. Validation of the research model...........................................................................
4.6.2. Research results.....................................................................................................
CHAPTER V: CONCLUSION AND RECOMMENDATION..............................................
5.1. Conclusion.....................................................................................................................
5.2. Recommendation..........................................................................................................
5.3. Limitation and Development.........................................................................................
REFERENCES......................................................................................................
APPENDIX..........................................................................................................

TABLES LIST
Table 1: Related research
Table 2: “Customization design” scale
Table 3: “Perceived Security” scale
Table 4: “Simple and convenient usage” scale
Table 5: “Awareness of BNPL” scale
Table 6: “User friendliness of BNPL” scale

3
Table 7: “Impulsive Buying Tendency” scale
Table 8: Descriptive statistics
Table 9: Table of results to evaluate the reliability of the variables
Table 10: Initial KMO and Bartlett's Test
a
Table 11: Initial Rotated Component Matrix
Table 12: KMO and Bartlett's Test
a
Table 13: Second Rotated Component Matrix
Table 14: KMO and Bartlett’s Test of Sphericity of Dependent variables
Table 15: Testing results of EFA of Dependent Variables
Table 16: Correlation matrix table
Table 17: Initial Model Summary
Table 18: ANOVA
Table 19: Coefficients
Table 20: Collinearity Statistics VIF
Table 21: Summary results of hypotheses testing

FIGURES LIST
Figure 1: Proposed research model
Figure 2: Research process
Figure 3: Percentage of people knowing BNPL feature
Figure 4: Percentage of people using BNPL feature
Figure 5: Percentage of genders using BNPL feature
Figure 6: Percentage of student level using BNPL feature
Figure 7: The Research Outcome Model
Figure 8: Normalized Residual Frequency Chart Histogram
Figure 9: Normalized Residual Frequency Chart Normal P-P Plot
Figure 10: Normalized Residual Frequency Chart Scatterplot

ABBREVIATIONS LIST
AD: Awareness of BNPL
Anova: Analysis of Variance
BNPL: Buy now, Pay later
CD: Customization Design
EFA: Exploratory Factor Analysis
IBT: Impulsive buying tendency
KMO: Kaiser – Meyer – Olkin
PS: Perceived Security

4
SC: Simple and Convenient usage
Sig: Observed Significance level
SPSS: Statistical Package for the Social Science

UF: User friendliness of BNPL

VIF: Variance inflation factor

5
 CHAPTER I: INTRODUCTION

1.1. Reason (Problem Statement)

In this modern world, with the swift evolution of e-commerce platforms, the feature of buy now
pay later is gradually being applied strongly on online shopping applications to help buyers in terms of
budget. Buy now, Pay later (BNPL) services offer convenience and flexibility to consumers by
enabling them to purchase items and make payments in installments., frequently with no interest
charged.. This can help individuals manage their budgets more effectively and afford larger purchases
they might not otherwise be able to afford upfront.
With the burgeoning expansion of digital finance, the increase in the adoption rate of BNPL as
a payment method is inevitable. This payment method is excessively utilized in online shopping or e-
commerce settings particularly (Arzaq Z. M, I Wayan N. L, [Link]., Ph.D., 2023). According to the
research findings, the value of BNPL transactions experienced explosive growth by around 55 percent
in 2019/20 and a threefold rise in total value over the whole period (Chay Fisher, Cara Holland, and
Tim West, 2021). That is because we are living in a growing inflationary society where everything
costs a bundle, people are now gravitating towards BNPL as this feature offers them the chance to buy
goods immediately without signing any physical contracts to satisfy their needs. People prefer using
BNPL features to credit cards as the BNPL charges them with no fees in a specific timeframe, no
physical contract, high security, no complicated procedures to register and huge amounts of loanable
funds.

However, behind the benefits received, as these offers motivate people to buy things when they
cannot afford, people usually overspend when shopping with BNPL. This leads to many unforeseen
consequences, specifically severe budget deficits resulting in an inability to repay debts, reduced
budget management capabilities, expenditure management, financial crises, etc. Moreover, the
simplicity and accessibility of BNPL services may contribute to heightened consumption and debt
accumulation, particularly GenZ. These warrant further investigation into the potential psychological
and behavioral of BNPL demographics, specifically exploring how it might influence impulsive buying
tendencies and long-term financial planning. Therefore, our team decided to research the factors in the
buy now pay later feature that influence students' overspending tendencies. We want to measure how
these factors impact and how strong they are, to propose useful and feasible solutions for students to
address this issue and stabilize and control their budgets. Considering various reasons mentioned

6
above, our study topic will be: “Factors contributing to the impulsive buying tendency of UEH
students using BNPL services.”

1.2. Aims of research

- Build the research model, identify factors of BNPL features that cause students to buy things
impulsively.

- Measure the degree of these factors on students’ impulsiveness on shopping.

- Suggest some viable solutions for students to control and balance their spending.

1.3. Research questions

- Question 1: Which factors of Buy now, Pay later motivate students to overconsume?

- Question 2: How do impulsive buying tendencies impact UEH students’ lives?

- Question 3: What recommendations do we propose for students to effectively monitor their spending?

1.4. Research methods

To address the problems raised in the topic, the authors implemented two main methods:
- Qualitative research: First the team developed an initial quantitative scale based on prior
research and projects which have similar trends to the research topic. Subsequently, following
consultations with the guiding instructor regarding the practical applicability of the research
topic, the entire team made revisions and enhancements to the quantitative scale to be included
in the questionnaire served for the quantitative research.
- Quantitative research: After the qualitative phase, the team developed a comprehensive
questionnaire with 23 Likert scale questions which employed a 5-point scale allowing the
sample surveyed to express various degrees of agreement or disagreement with specific
statements. The data gathered was processed using the SPSS software to assess the scale's
reliability through Cronbach's Alpha coefficients. Then, Exploratory Factor Analysis and
multiple regression analysis were carried out to analyze the effect of each element concluding
with the evaluation of the team's proposed model and hypothesis testing.

7
 CHAPTER II: THEORETICAL BASIS

2.1. Related research

Delving deeper into the issue of factors affecting UEH students' impulsive shopping tendency when
using BNPL services, especially with the keywords BNPL, impulsive shopping, the group found a
number of articles and papers. related research. These articles are summarized in Table 1.

Method/Scale of
No Name of research Content of research
research

Determine elements that

“FACTORS AFFECTING have the impact on the
211 respondents ( from
INTENTION TO USE OF ONLINE tendency to embrace pay
18 to 24 years old)
1 INSTALLMENT PAYMENT later services using a
gathered from Google
PAYLATER” (Ivonanda Nindya and framework derived from
Form
Rifelly D. A, 2022) the diffusion of innovation
theory.

18,887 people who got

into bankruptcy or
entered into a debt
“THE ROLE OF CREDIT CARDS,
agreement between
PAYDAY LOANS, CONSUMER Examine the impact of
July 1, 2016, and June
LEASES AND BUY-NOW-PAY-LATER payday loans, consumer
2 30, 2019 have their
PRODUCTS IN PERSONAL leases, and BNPL debt on
debts listed in the
INSOLVENCY” (Lucinda O’Brien, Ian personal insolvency.
statistics.
Ramsay and paul ali, 2022)

3 “Click to Buy: The Impact of Retail The research aims at Investigate how young
Credit on Over-Consumption in the collecting data from adult female purchasers
Online Environment” (Fook & 109 females that age doing online fashion
McNeill, 2020) between 18 and 25. shopping among is
influenced by a range of
impulsive enabling

8
 financial tools (BNPL
credit schemes)

Multi-stage (Selection
of review participants, Develop a broad
Information requests, understanding of BNPL
“Review of buy now pay later
4 Comparison with credit arrangements and to
arrangements” by ASIC, 2018
cards, Stakeholder identify potential risks for
consultation, Consumer consumers.
research)

533 individuals using

BNPL participated in The psychological factors
“Understanding the Psychological
story stem completion. driving BNPL usage in the
Determinants of Buy-Now-Pay-Later
Template analysis was UK are explored, and the
5 (BNPL) in the UK: A User
performed and effectiveness of current
Perspective” (Taylor et al., 2015)
identified four BNPL payment structures is
sentiment groups for assessed.
comparison.

Table 1: Related research

2.2. Literature review

2.2.1. Buy Now, Pay Later

Buy Now, Pay Later (BNPL), is a payment method widely used for e-commerce settings, offers
convenience and flexibility to consumers by allowing them to make any purchases and pay in
installments, without interest if paid on a set timeframe. BNPL services and existing installment-based
payment methods are distinguished by two key distinctions. Firstly, BNPL services are administered by
third-party entities rather than directly from the seller from which the customers buy goods. Moreover,
most BNPL services impose fees on merchants, deducting a percentage of each sale before transmitting
the remaining balance to the merchant. Additionally, BNPL services are utilized online., with their
engagement through dedicated websites or mobile apps. These platforms facilitate the application
process for determining a spending limit, likely similar to the traditional process of establishing a credit
limit. The amount enabled for customers to borrow depends on creditworthiness, the specific BNPL
provider, and the individual’s financial situation. While BNPL services are primarily utilized for online

9
purchasing, its adaptation to physical retail is also functioned through their apps. The apps generate
scannable QR codes, or virtual cards, both of which can be used for instore and online transactions
(Cook et la., 2023). Another difference between BNPL and the typical lay-by method is that customers
receive their purchases immediately and the merchant is paid up front by the BNPL service provider
(Chay Fisher, Cara Holland, and Tim West, 2021). BNPL is available from a variety of companies,
ranging from established financial businesses, technology firms, e-commerce platforms. Some of the
most popular BNPL service providers are Afterpay, PayPal credit, Shopee. The facilities of the services
might slightly differ depending on the provider terms and conditions.

2.2.2. Impulsive buying tendency

A phenomenon when consumers make purchases without any prior planning or consideration is
defined as impulsive buying tendency. This definition by Stern (1962) has laid the foundation for much
of the subsequent research on impulse buying tendency. Rook (1987) described this tendency as “a
consumer experiences a sudden, often powerful and persistent urge to buy something immediately”.
Similarly, impulse buying involves making rapid, unplanned decisions driven by a strong urge to
possess an item immediately (Rook & Gardner, 1993). Relying on the definitions provided, impulse
buying tendency can be described as a spontaneous purchasing tendency that occurs when a shopper is
exposed to a stimulus and experiences a sudden urge to make a purchase, and it is usually more
emotional than rational.

Previous studies have also indicated the influence of BNPL on customers’ impulsive shopping
tendency. Hirschman (1979) highlighted the significant impact of payment methods on how people
shop. He pointed out that installment plans, often enabled by credit cards, allowed consumers to
purchase expensive goods without needing immediate full payment. Further research by Prelec &
Loewenstein (1998) and Soman (1998) suggests that zero-interest payment plans can trigger impulsive
buying tendency in consumers. As breaking down the cost into smaller payments, items seem to be less
expensive upfront. The ease of spreading payments over time can drive consumers to make purchases
without considering future financial burdens. Consumers who shop without a plan are more likely to
result in post-purchase regret (Saleh, 2012).

2.2.3. Interface Design

Humans are naturally drawn to things that are visually appealing and stand out. An app with a beautiful
interface, harmonious colors, and a logical layout will be more likely to attract users' attention than an
app with a monotonous and boring interface. Kotler (1973) said “the conscious designing of space to

10
create certain buyer effects.” An app with a beautiful and easy-to-use interface will be more
aesthetically pleasing to users and will make them more likely to want to use it.

2.2.4. Perceived Control

In terms of psychology, perceived control (PC) refers to A person's personal confidence in their
capabilities. to influence themselves, their surroundings (including people, places, and things), and
their own emotional state and activities. A sense of perceived control is positively correlated with
emotional well-being, a diminished physiological response to stressors, improved stress management
skills, enhanced performance, reduced pain perception, and an increased propensity for successful
implementation of challenging behavioral modifications. (Thompson & Spacapan, 1991)

2.2.5. Perceived Security

In the digital technology age, the development of online shopping raises the concern about
security. At the heart of this concern lies the concept of "perceived security" (Flavián & Guinalíu,
2006). This term reflects a user's subjective belief that their personal information – from account details
to transaction history – remains safe from unauthorized access, storage, or manipulation. Unfortunately,
the specter of hacking looms large, leading many consumers to perceive digital payment systems as
inherently risky (Falk et al., 2016). This perception of vulnerability emerges as one of the most
significant hurdles to wider adoption of these technologies (C. Kim et al., 2010). Hence, building trust
and confidence in the security measures employed by digital payment platforms is crucial to ensure
their continued growth and success.

2.3. Research Hypotheses

Aim of research: Identify factors that contribute to the impulsive buying tendency of students using
BNPL services.

2.3.1. Dependent variable: Impulsive Buying Tendency

The BNPL initiative represents a recent addition to the consumer retail landscape, positioning itself in
direct competition with established conventional lending practices (Edmunds, 2018). The BNPL
payment option has increasingly influenced consumer purchasing behaviors, enabling consumers to
make immediate purchases while deferring payments, often with little to no interest. Research suggests
that BNPL services can lead to higher spending and impulsive buying tendencies due to reduced
immediate financial constraints (King, 2021).

The BNPL service has been shown to significantly influence consumer decision-making, often
encouraging impulsive purchases due to its flexibility and deferred payment structure. The availability

11
of BNPL options increased the likelihood of consumers making unplanned purchases, particularly for
higher-ticket items, by reducing the immediate psychological cost of spending (Miller and Roberts,
2020).

Therefore, the author group decided to conduct a scientific research to test whether the BNPL features
have impacts on the customers’ impulsive buying tendency. According to these, the dependent variable
Y is proposed as follows:

“Dependent variable Y: Customers’ impulsive buying tendency is influenced by Customization design,

Perceived Security, Simple and Convenient Usage, User’s Awareness and User Friendliness.”

2.3.2. The impact of BNPL’s Customization Design on users’ impulsive buying tendency
Customization on a mobile app or service, one of the most essential features, is a process of
allowing users to modify or adjust its features, interface and functionality according to their needs and
preferences (Jianxing, Yan & Yao, 2019). This is because users spend more time on interface overtime,
their needs evolve dramatically, their proficiency in using it increases and then alter it to a more
advanced version. Consequently, customization has the potential to improve users’ experience using
the service, which is thought to impact various psychological consequences (Sampada Marathe & S.
Shyam Sundar, 2011). This can result in a higher number of impulsive purchases as the time used for
browsing the services and with the more personalized settings, the more they catch items suitable with
their interests. According to these, the hypothesis H1 is proposed as follows:

“H1: The Customization Design has a positive (+) impact on the impulsive buying tendency among
UEH students”

2.3.3. The impact of BNPL’s Perceived Security on users’ impulsive buying tendency
The sense of security regarding online payments refers to the extent to which a user believes that
making transactions on online platforms are secure, covering both financial and personal information
(Hartono et al., 2014). Nowadays, the issue of purchasers’ information security is highly concerning,
especially as online payments are becoming increasingly popular. Customers who experience a lack of
security when using BNPL services may cause them to become anxious and more reluctant to use
them. Consequently, customer trust in a service provider's security measures directly correlates with
their willingness to use BNPL (Agustin, 2022). When customers believe that a service provider has
strong security measures in place to prevent unauthorized activities and protect their data, they are
more likely to use BNPL services. The importance of perceived security in influencing purchase
intention is highlighted since customers only engage in online shopping when they believe that their

12
personal information is secure, (Salisbury et al., 2001). Derive from this, the hypothesis H2 is proposed
as follows:

“H2: The perceived security of BNPL has a positive (+) impact on the impulsive buying tendency of
UEH students”

2.3.4. The impact of BNPL’s Simple and Convenient usage on users’ impulsive buying tendency
BNPL offers a convenient, flexible payment option where there has not been one before for many
consumers (Reed Hartman, 2021). People find BNPL convenient because of its user-friendly interface
and seamless integration with payment gateways and online retailers, providing a smooth shopping
experience. Typically, users select the BNPL option at checkout, provide basic personal information,
and confirm their payment plan (Amy Boyington, 2024). BNPL allows users to create an account and
securely store their personal information for future transactions, eliminating the need to repeatedly
enter the same details for each purchase, saving time and simplifying the payment process. Given these
advantages, we propose the following hypothesis H3:

“H3: Simple and convenient usage has a positive (+) impact on the impulsive buying tendency of UEH
students”

2.3.5. The impact of BNPL’s Awareness on users’ impulsive buying tendency

Lots of users do not realize that it is a form of credit, the amount of money they have to pay is
not reduced, it is just divided into smaller ones. Therefore, consumers could not recognize their real
financial situation. “It’s important to recognize that deferring payments and thinking of it as ‘next
month’s problem’ can lead to a cycle of debt” (Thompson C., 2023). However, if buyers understand
about this trap, they would be aware of their paying ability and have their own effective financial
management. According to these, the hypothesis H4 is proposed as follows:
“H4: Awareness of BNPL has a negative (-) impact on the impulsive buying tendency of UEH
students"

2.3.6. The impact of BNPL’s User-Friendliness on users’ impulsive buying tendency

One significant reason why online customers are reluctant to shop online is the absence of trust
between the majority of customers and businesses and in e-commerce (Grabner-Kraeuter S., 2002).
Implementing purchase policies can address this issue, enhancing customer trust and creating a more
secure online shopping environment. BNPL services often offer a variety of friendly policies, such as
convenient product return policies, flexible exchange policies, etc., along with the clear and
straightforward information about its services, fees, terms and conditions, and refund options, future
payments. When consumers perceive that the risks associated with online transactions are reduced and
they can have better control over their purchases in the online environment, they are more willing to

13
engage in transactions (Pavlou P., 2001). Aligning guarantee policies with certifications from
independent third parties can also streamline the process of earning and maintaining consumer trust
(Grabner-Kraeuter , 2002). According to these, the hypothesis H5 is proposed as follows:
“H5: User friendliness of BNPL has a positive (+) impact on the impulsive buying tendency of UEH
students.”

14
2.4. Proposed Research Model

All of the proposed hypotheses that have a relation to the outcome variable have been visualized as in
the Figure 1:

Customization design
(H1)

H1 (+)
Perceived Security H2 (+)
(H2)

H3 (+) Impulsive Buying Tendency

Simple and
convenient usage (H3) (Y)
H4 (-)

Users’ Awareness H5 (+)

(H4)

User friendliness
(H5)

Figure 1: Proposed research model

15
 CHAPTER III: METHODOLOGY

3.1. Research process

The overview of main steps in the research procedure is illustrated in Figure 2.

Figure 2: Research process

16
3.2. Research subject and scope

3.2.1. Research sample

● Survey subjects: Students of UEH University from 46, 47, 48, 49 batches.

3.2.2. Research scope

● Time: Our group conducted the project in the first semester of 2024, from April to the end of

May.

● Space: We conduct research at all UEH campuses in Ho Chi Minh city.

3.3. Research methods

To address the problems raised in the topic, the authors implemented two main methods:

3.3.1. Qualitative research

First, the team developed an initial quantitative scale based on prior research and projects which
have similar trends to the research topic. Building on the inheritance from previous research, the
authors employed a research model to analyze the factors causing users’ impulsive buying tendencies
when using BNPL.
Subsequently, following consultations with the guiding instructor regarding the practical
applicability of the research topic, the entire team made revisions and enhancements to the quantitative
scale to be included in the questionnaire used for the quantitative research.

3.3.2. Quantitative research

After the qualitative phase, the team developed a comprehensive questionnaire with 23 Likert
scale questions which employed a 5-point scale allowing the sample surveyed to express various levels
of agreement or disagreement with each specific statement. The author group employed an online
survey approach to gather and refine data, constructing a quantitative dataset to support the analysis
using the Google Forms platform.
The gathered data was processed using the SPSS software to evaluate the scale’s reliability
through Cronbach's Alpha coefficients, allowing the team to assess and eliminate unsuitable variables.
Then, EFA and regression analysis were carried out to analyze the impact of each factor, concluding
with the evaluation of the team's proposed model and hypothesis testing.

17
3.4. Sample size

This study will carried out Exploratory Factor Analysis and Multiple Regression Analysis test.
To determine the sample size, rely on the previous studies, it can be seen that the minimum sample size
for conducting EFA and multiple regression analysis follows the principles below:
For EFA, based on the research work of Hair, Anderson, Tatham, and Black published in 1998,
the minimum requirement for a sample size is calculated by the formula n = 5*m (in which m is the
number of observed variables). This research has a total of 19 observed variables, so the minimum
sample size will be n = 95. For multiple regression analysis, in accordance with the theory of
Tabachnick and Fidell presented in their 1996 work, the minimum sample size is measured by the
equation n = 50 + 8*m (where m is the number of independent variables). With 5 independent variables
in this research, the minimal sample size for multiple regression analysis is n = 90. Therefore, the
authors determined that the quantitative investigation would require a minimum sample size of 175
participants.

3.5. Sampling technique

The research team chose the convenient sampling method as we proceeded data collected from
the online forms sent to UEH students as well as remove responses that do not fall within the target
group that the team wanted to survey, while also eliminating any outliers in the study. After filtering
the data, only 175 responses met the requirements.

3.6. Procedure of collecting data

This approach is commonly used in scientific fields by researchers, which involves gathering
data from relevant and reliable papers and conducting their own surveys. The authors of this study also
do the same to understand how "Buy now, pay later" options affect students' impulse buying habits.
They also crafted a questionnaire and targeted UEH students to gather their thoughts and insights into
this topic.

18
3.7. Measurement Scales

3.7.1. “Customization design” scale

Customization Design – H1 (+)

Code Original Items Adapted Items In Vietnamese Sources

CD1 “I feel that I can I feel that I can Tôi cảm thấy rằng tôi 1. Anh et al(2023)
customize the customize the có thể tùy chọn các
2. Jiaxin Zhang,
payment methods payment methods phương thức thanh
Yan Luximon and
according to my own according to my own toán theo mong muốn
Yao Song (2019)
preference”. preference. của mình.

CD2 I feel that I can I feel that I can Tôi nhận thấy rằng
customize the privacy customize the privacy tôi có thể đảm bảo sự
settings according to settings according to riêng tư theo mong
my own preference. my own preference. muốn mình.

CD3 I feel that I can I feel that I can Tôi cảm thấy rằng tôi
customize the customize the có thể cài đặt bảo mật
security settings security settings theo mong muốn
according to my own according to my own mình
preference preference

Table 2: “Customization design” scale

19
3.7.2. “Perceived Security” scale

Perceived Security – H2 (+)

Code Original Items Adapted Items In Vietnamese Sources

PS1 I perceive security I perceive security Tôi nhận thức được 1. Anh et al.
using my credit/debit using my BNPL sự bảo mật khi sử (2023)
card information information through dụng thông tin BNPL
2. Jiaxin Zhang,
through mobile mobile payment của mình thông qua
Yan Luximon and
payment platforms. platforms. nền tảng thanh toán
Yao Song (2019)
di động.

PS2 I would feel safe I would feel safe Tôi sẽ cảm thấy an
providing sensitive providing sensitive toàn khi cung cấp
information about information about thông tin nhạy cảm
myself over the myself over the về bản thân trên nền
mobile payment BNPL platforms. tảng BNPL.
platforms.

PS3 I perceive that mobile I perceive that BNPL Tôi hiểu rằng nền
payment platforms platforms are secure tảng BNPL là hệ
systems to conduct a thống an toàn để thực
are secure systems to
transaction. hiện giao dịch.
conduct a transaction.

Table 3: “Perceived Security” scale

20
3.7.3. “Simple and convenient usage” scale

Simple and convenient usage – H3 (+)

Code Original Items Adapted Items In Vietnamese Sources

SC1 BNPL is very I feel that BNPL is Tôi cảm thấy BNPL Upendra
convenient and very convenient and rất tiện lợi và có sẵn Goonetilleke
available at available at your trong tầm tay. (2022)
fingertips. fingertips.

SC2 You need to enter I find that BNPL Tôi thấy rằng BNPL
personal details only needs to enter chỉ cần nhập thông
once. personal details only tin cá nhân một lần.
once.

SC3 Excellent customer I find that BNLP has Tôi thấy rằng BNPL
service of BNPL excellent customer có dịch vụ khách
which gives a service, which gives a hàng tuyệt vời mang
welcoming feeling. welcoming feeling. lại cảm giác thân
thiện.

SC4 Very clear and I find that BNPL has Tôi thấy BNPL có hệ
transparent system a very clear and thống rất rõ ràng,
with regular transparent system minh bạch với những
reminders and with regular lời nhắc nhở, tuyên
statements. reminders and bố thường xuyên.
statements.

Table 4: “Simple and convenient usage” scale

21
3.7.4. “Awareness of BNPL” scale

Awareness of BNPL – H4 (-)

Code Original Items Adapted Items In Vietnamese Sources

AD1 Easy to buy unwanted I find it hooked to Tôi cảm thấy cuốn Upendra
items with BNPL. buy unwanted items hút khi mua những Goonetilleke
with BNPL. món không cần thiết (2022)
khi sử dụng BNPL.

AD2 The tendency to buy I tend to buy things Tôi có xu hướng mua
things people cannot that I cannot afford những thứ mà tôi
afford due to BNPL. due to BNPL. không đủ khả năng
chi trả do BNPL.

AD3 The appetizing nature I am usually unaware Tôi thường không

of BNPL makes it an of the appetizing quan tâm đến sự thúc
ignorant usage. nature of BNPL. đẩy tiêu dùng của
tính năng BNPL.

AD4 Not consciously I do not need to save Tôi cảm thấy không
thinking about saving money because a cần tiết kiệm lại sau
money as a portion is portion is already khi chi một khoản
already paid. paid when purchasing tiền để mua sắm bắng
with BNPL. BNPL.

Table 5: “Awareness of BNPL” scale

22
3.7.5. “User friendliness of BNPL” scale

User friendliness of BNPL – H5 (+)

Code Original Items Adapted Items In Vietnamese Sources

UF1 Convenient product I think the product Tôi thấy chính sách Upendra
return policies. return policies of trả sản phẩm của Goonetilleke
BNPL are BNPL rất thuận tiện. (2022)
convenient.

UF2 Flexible exchange I think BNPL has Tôi nghĩ BNPL có

policies. flexible exchange chính sách đổi sản
policies. phẩm linh hoạt.

UF3 Future installments When using BNPL, I Khi sử dụng BNPL,

are known from the find that BNPL offers tôi thấy rằng BNPL
start itself. complete cung cấp sự minh
transparency about bạch hoàn toàn về các
your future khoản trả góp trong
installments from the tương lai của bạn
start. ngay từ đầu.

UF4 Transparent public I find that BNPL Tôi thấy rằng BNPL
platform of BNPL. offers a high level of cung cấp mức độ
transparency on their minh bạch cao trên
public platform. nền tảng công khai
của họ.

Table 6: “User friendliness of BNPL” scale

23
3.7.6. “Impulsive Buying Tendency” scale

Impulsive Buying Tendency – Y

Code Original Items Adapted Items In Vietnamese Sources

IBT1 Most of my purchases I do not plan in Tôi không lên kế Talaat Abo
are planned in advance what to hoạch trước những gì Elmaaty Ibrahim
advance. purchase. sẽ mua. Ahmed (2023)

IBT2 Before I buy I find it unnecessary Tôi thấy không cần

something I always to consider what to thiết phải cân nhắc
carefully consider buy when shopping nên mua gì khi dùng
whether I need it. with BNPL. BNPL.

IBT3 I sometimes buy I usually make Tôi thường mua hàng

things because I like purchases based on dựa trên sở thích hơn
buying things, rather desire rather than là độ hữu dụng.
than because I need necessity.
them.

IBT4 I buy what I like I tend to make Tôi có xu hướng mua

without thinking impulsive purchases sắm bốc đồng mà
about the without considering không nghĩ đến hậu
consequences. the consequences. quả.

IBT5 I ended up spending I ended up spending Sau cùng, tôi đã chi

more money on my more money on my nhiều tiền hơn cho
purchases than I purchases than I việc mua sắm so với
originally set out to intended to. dự định ban đầu.
spend.

Table 7: “Impulsive Buying Tendency” scale

24
3.8. Data analysis method
Since we have the result from the Google Form, we recheck the data and upload it to the SPSS.25 for
further processing and data analysis. Specifically, as follow:

3.8.1. Descriptive Statistics

- Descriptive statistics for categorical variables: Research and evaluate the statistical data about
genders, school years, awareness and usage of BNPL level of survey participants.
- Descriptive statistics for observed variables: Research and calculate maximum and minimum value,
average, median and standard deviation of observed variables.

3.8.2. Measure Reliability Assessment

Cronbach’s Alpha coefficient: using for conducted a scale reliability test and eliminate the unreliable
observed variables which based on these criteria:
- Examining each observed variable group of each factor.
- Scale’s Cronbach’s Alpha value that higher than 0.6 is acceptable.
- Regarding the scale's Cronbach's Alpha coefficient being less than 0.6, those variables are
removed in order to maximize the Cronbach’s Alpha value or Cronbach’s Alpha if Item Deleted
value and continue recalculated until the scale’s Cronbach’s Alpha coefficient is at least 0.6 or
higher.
- Eliminating the variables has Corrected Item – Total Correlation smaller than 0.4.

3.8.3. Exploratory Factor Analysis

Exploratory factor analysis (EFA) is a statistical technique employed to condense a large group
of related observed variables into a more manageable set of unobserved latent variables or factors. EFA
is a data-driven technique that does not require a priori hypotheses about the number or nature of the
underlying factors (Hair et al. 2009).
If the Cronbach's Alpha scale does not consider the relationships between all observed variables in
different factors, EFA examines the relationships between variables in all different groups (factors) to
detect variables that load on multiple factors or variables that are misclassified from the beginning
based on the following criteria:

● 0.5 ≤ KMO (Kaiser-Meyer-Olkin) ≤ 1.

25
 ● Bartlett's test has a Sig < 0.05 (with H0: variables are not correlated with each other in the

population, H1: variables are correlated with each other in the population).

● Total Variance Explained in factor extraction > 50%.

● Eigenvalue ≥ 1 (Eigenvalue is a criterion for determining the number of factors in EFA. The

minimum Eigenvalue must be equal to 1 with this criterion).

● Factor loading > 0.4.

3.8.4. Pearson Correlation Analysis

In quantitative research using SPSS, conducting a Pearson correlation is a crucial initial step.
This analysis helps identify strong linear relationships between dependent and independent variables,
as well as detect potential multicollinearity when independent variables are highly correlated.
According to Andy Field (2009), to assess the validity of the Pearson correlation coefficient, its
statistical significance must be tested. The null hypothesis (H0: r = 0) suggests no correlation, and a t-
test is used to evaluate this hypothesis.
If the significance value is less than 0.05, the null hypothesis is rejected, indicating a
statistically significant non-zero correlation (r ≠ 0) and a linear relationship between the variables.
Conversely, if the significance value exceeds 0.05, the null hypothesis is retained, suggesting no
statistically significant correlation (r = 0) and no linear relationship.

3.8.5. Multivariate Regression Analysis

Multivariate regression analysis is a key method in data analysis, used to examine the linear
relationships between multiple independent variables and multiple dependent variables. Once the
correlations between the independent and dependent variables have been established, multivariate
regression analysis is performed to provide a more detailed understanding of how the independent
variables influence the dependent variables. The regression model defines the nature of these
relationships, allowing for the prediction of the dependent variables' values based on the known values
of the independent variables.

26
 CHAPTER IV: RESEARCH RESULTS
4.1. Descriptive statistics of the survey sample

The author team carried out an online survey on the target audience and collected 233 answers. After
filtering the data, we removed 28 responses that state they did not know about BNPL before and 30
answers state that they have not used this paying method yet.

Figure 3: Percentage of people knowing BNPL feature

Figure 4: Percentage of people using BNPL feature

27
At last, only 175 responses fulfilled the conditions input into SPSS software for data analysis. The
statistical results are summarized in tables with the following specific details:

Figure 5: Percentage of genders using BNPL feature

Figure 6: Percentage of student level using BNPL feature

- About gender: There were 98 male respondents corresponding to 56% and the rest were female
(77 female for 44%).

28
- About school year: The proportion of respondents are freshman comprise the largest number in
the survey (55,4%), there were 26,9% and 17,7% of respondents are sophomore and junior
respectively.

29
4.2. Descriptive statistics of the independent variables

Std.
Code Min Max Mean Variance
Deviation
Customization Design H1

I feel that I can customize the CD1 1 5 3.73 1.142 1.304

payment methods according to my
own preference.
I feel that I can customize the CD2 1 5 3.37 1.238 1.532
privacy settings according to my
own preference.
I feel that I can customize the CD3 1 5 3.62 1.101 1.213
security settings according to my
own preference.
Perceived Security H2

I perceive security using my BNPL PS1 1 5 3.47 1.087 1.181

information through mobile
payment platforms.
I would feel safe providing PS2 1 5 3.10 1.153 1.330
sensitive information about myself
over the BNPL platforms.
I perceive that BNPL platforms are PS3 1 5 3.47 1.071 1.147
secure systems to conduct a
transaction.
Simple and Convenient usage H3

SC1 1 5 3.72 1.153 1.329

I feel that BNPL is very convenient
and available at fingertips.

SC2 1 5 3.61 1.198 1.434

I find that BNPL needs to enter
personal details only once.

30
I find that BNLP has excellent SC3 1 5 3.42 1.116 1.245
customer services which gives
welcoming feelings.
I find that BNPL has very clear and SC4 1 5 3.59 1.115 1.242
transparent system with regular
reminders and statements.
Awareness of BNPL H4

AD1 1 5 3.13 1.208 1.460

I find it hooked to buy unwanted
items with BNPL.

AD2 1 5 3.05 1.261 1.589

I tend to buy things that I cannot
afford due to BNPL.

AD3 1 5 3.34 1.216 1.479

I am usually unaware of the
appetizing nature of BNPL.

I do not need to save money AD4 1 5 3.07 1.276 1.627

because a portion is already paid
when purchasing with BNPL.
User friendliness of BNPL H5

I think the product return policies UF1 1 5 3.57 1.101 1.212

of BNPL are convenient.

I think BNPL has flexible UF2 1 5 3.52 1.093 1.194

exchange policies.

When using BNPL, I find that UF3 1 5 3.59 1.136 1.290

BNPL offers complete
transparency about your future
installments from the start.
I find that BNPL offers a high level UF4 1 5 3.67 1.111 1.234
of transparency on their public

31
 platform.
Impulsive Buying Tendency Y

IBT1 1 5 2.95 1.332 1.773

I do not plan in advance what to
purchase.

I find it unnecessary to consider IBT2 1 5 2.58 1.265 1.601

what to buy when shopping with
BNPL.
IBT3 1 5 2.95 1.224 1.497
I usually make purchases based on
desire rather than necessity.

I tend to make impulsive purchases IBT4 1 5 2.86 1.290 1.663

without considering the
consequences.
I ended up spending more money IBT5 1 5 3.31 1.277 1.631
on my purchases than I intended to.

Table 8: Descriptive statistics

After running SPSS with the data gathered from the sample surveyed, the statistical results
revealed that most of the mean of the independent variables is greater than 3.1/5, with a standard
deviation ranging from 1.07 to 1.28. The observed values range from a minimum of 1 to a maximum of
5. We observed that many opinions were agreed upon by the respondents, with the observed variables
related to BNPL awareness having the lowest average values (ranging from 3.05 to 3.34).
In the dependent variable measuring Impulsive Buying Tendency (IBT), the majority of the
responses received showed disagreement, with most of the mean value being lower than 3, ranging
from 2.58 to 2.95. Among these, IBT5 perception had the highest mean value at 3.31. We perceived
that most people think that BNPL (Buy Now, Pay Later) does not cause impulsive buying tendencies;
however, it turns out that BNPL causes them to shop more than they intended to. Given the uncertainty
surrounding the precise effects of BNPL factors on impulsive buying tendencies, we intend to employ
regression analysis to determine which factors are statistically significant and to better understand their
individual influences on user behaviors when utilizing BNPL services.

32
4.3. Cronbach’s Alpha: Scale Reliability Test
The results of the Cronbach's Alpha analysis for the scales are summarized in the following tables:

Observable Variables Corrected Conclusion

Item-Total
Correlation
Customization Design (H1)

Cronbach's Alpha = .799

CD1 - I feel that I can customize the payment methods .634 Suitable
according to my own preference. Variables
CD2 - I feel that I can customize the privacy settings according .693 Suitable
to my own preference. Variables

CD3 - I feel that I can customize the security settings according .610 Suitable
to my own preference. Variables

Perceived Security (PS)

Cronbach's Alpha = .756

PS1 - I perceive security using my BNPL information through .572 Suitable

mobile payment platforms. Variables
PS2 - I would feel safe providing sensitive information about .637 Suitable
myself over the BNPL platforms. Variables

PS3 - I perceive that BNPL platforms are secure systems to .549 Suitable
conduct a transaction. Variables

Simple and Convenient Usage (SC)

Cronbach's Alpha = .720

33
SC1 - I feel that BNPL is very convenient and available at .614 Suitable
fingertips. Variables
SC2 - I find that BNPL needs to enter personal details only .530 Suitable
once. Variables

SC3 - I find that BNLP has excellent customer services which .582 Suitable
gives welcoming feelings. Variables

SC4 - I find that BNPL has a very clear and transparent system .326 Not Suitable
with regular reminders and statements. Variables
Awareness of BNPL (AD)
Cronbach's Alpha = .724

AD1 - I find it hooked to buy unwanted items with BNPL. .557 Suitable
Variables
AD2 - I tend to buy things that I cannot afford due to BNPL. .609 Suitable
Variables

AD3 - I am usually unconscious of the appetizing nature of .350 Not Suitable

BNPL. Variables

AD4 - I do not need to save money because a portion is already .545 Suitable
paid when purchasing with BNPL. Variables
Users’ Friendliness (UF)

Cronbach's Alpha = .788

UF1 - I think the product return policies of BNPL are .623 Suitable
convenient. Variables
UF2 - I think BNPL has flexible exchange policies. .598 Suitable
Variables

34
 UF3 - When using BNPL, I find that BNPL offers complete .672 Suitable
transparency about your future installments from the start. Variables

UF4 - I find that BNPL offers a high level of transparency on .493 Suitable
their public platform. Variables
Impulsive Buying Tendency
Cronbach's Alpha = .845

.695 Suitable
IBT1 - I do not plan in advance what to purchase.
Variables
.671 Suitable
IBT2 - I find it unnecessary to consider what to buy when
Variables
shopping with BNPL.

.654 Suitable
IBT3 - I tend to make impulsive purchases without considering
Variables
the consequences.

IBT4 - I find that BNPL offers a high level of transparency on .646 Suitable
their public platform. Variables
.591 Suitable
IBT5 - I ended up spending more money on my purchases than
Variables
I intended to.

Table 9: Table of results to evaluate the reliability of the variables

Cronbach’s Alpha results of the scale “Customization Design” are 0.799 > 0.6; the corrected item-
Total Correlation coefficients of the total variables of the observable variables of the scale are all
greater than 0.4. Therefore, all the observable variables are acceptable and will be used in further factor
analysis.
Cronbach’s Alpha results of the scale “Perceived Security” are 0.756 > 0.6; the corrected item-Total
Correlation coefficients of the total variables of the observable variables of the scale are all greater than
0.4. Therefore, all the observable variables are acceptable and will be used in further factor analysis.
Cronbach’s Alpha results of the scale “Simple and Convenient Usage” are 0.720 > 0.6; the corrected
item-Total Correlation coefficients of the total variables of the observable variables of the scale are all

35
greater than 0.4, except for SC4. Therefore, all the observable variables, excluding SC4, are acceptable
and will be used in further factor analysis.
Cronbach’s Alpha results of the scale “Awareness of BNPL” are 0.720 > 0.6; the corrected item-Total
Correlation coefficients of the total variables of the observable variables of the scale are all greater than
0.4, except for AD3. Therefore, all the observable variables, excluding AD3, are acceptable and will be
used in further factor analysis.
Cronbach’s Alpha results of the scale “User Friendliness of BNPL are 0.724 > 0.6; the corrected item-
Total Correlation coefficients of the total variables of the observable variables of the scale are all
greater than 0.4. Therefore, all the observable variables are acceptable and will be used in further factor
analysis.
Cronbach’s Alpha results of the scale “Impulsive Buying Tendency” are 0.845 > 0.6; the corrected
item-Total Correlation coefficients of the total variables of the observable variables of the scale are all
greater than 0.4. Therefore, all the observable variables are acceptable and will be used in further factor
analysis.
After conducting the Cronbach’s Alpha test, the authors perceived that there were 2 observed variables
that have negligible total correlation (SC4 and AD3). Therefore, in the new independent variables, the
SC4 and AD3 are excluded. This will help running the Bivariate test be more accurate.

4.4. Exploratory Factor Analysis: Scale Validity Test

After being measured under Cronbach’s Alpha method to show how reliable the scales are, they
are now running under the EFA method to measure how well the internal structure of the scales are.
The Cronbach’s Alpha results show that there are a total of 18 observed variables of 5 independent
factors qualified enough to run the EFA test. The details are as follows:
“KMO and Bartlett's Test”
Kaiser-Meyer-Olkin Measure of Sampling .862
Adequacy.
Bartlett's Test of Approx. Chi-Square 1431.749
Sphericity
df 153

Sig. .000

Table 10: Initial KMO and Bartlett's Test

“Rotated Component Matrixa”

36
 Component

1 2 3 4 5
CD1 - I feel that I can customize the .419 .429 .484
payment methods according to my own
preference.
CD2 - I feel that I can customize the .623
privacy settings according to my own
preference.
CD3 - I feel that I can customize the .645 .486
security settings according to my own
preference.
PS1 - I perceive security using my BNPL .685
information through mobile payment
platforms.
PS2 - I would feel safe providing .757
sensitive information about myself over
the BNPL platforms.
PS3 - I perceive that BNPL platforms are .514 .584
secure systems to conduct a transaction.
SC1 - I feel that BNPL is very .798
convenient and available at fingertips.
SC2 - I find that BNPL needs to enter .837
personal details only once.
SC3 - I find that BNLP has excellent .466
customer services which gives
welcoming feelings.
SC4 - I find that BNPL has very clear .457 .573
and transparent system with regular
reminders and statements.
AD1 - I find it hooked to buy unwanted .648
items with BNPL.

37
 AD2 - I tend to buy things that I cannot .727
afford due to BNPL.
AD3 - I am usually unconscious of the .743
appetizing nature of BNPL.
AD4 - I do not need to save money .821
because a portion is already paid when
purchasing with BNPL.
UF1 - I think the product return policies .754
of BNPL are convenient.
UF2 - I think BNPL has flexible .808
exchange policies.
UF3 - When using BNPL, I find that .601
BNPL offers complete transparency
about your future installments from the
start.
UF4 - I find that BNPL offers a high .588
level of transparency on their public
platform.

Table 11: Initial Rotated Component Matrixa

Using the Principal Component Analysis method with Varimax rotation when analyzing factors
for 18 observed variables. The results of the initial EFA show that the Kaiser-Meyer-Olkin (KMO)
measure is 0.862 > 0.5; Bartlett's Test of Sphericity has a significance of 0.000 < 0.05; the total
variance explained is 66.098% > 50%; and all Eigenvalues are greater than 1. However, there are 8
observed variables that do not converge:
CD1 (I feel that I can customize the payment methods according to my own preference);
CD2 (I feel that I can customize the privacy settings according to my own preference);
CD3 (I feel that I can customize the security settings according to my own preference);
PS3 (I perceive that BNPL platforms are secure systems to conduct a transaction);
SC3 (I find that BNLP has excellent customer services which gives welcoming feelings);
SC4 (I find that BNPL has very clear and transparent system with regular reminders and statements);

38
AD3 (I am usually unconscious of the appetizing nature of BNPL);
UF4 (I find that BNPL offers a high level of transparency on their public platform)

After removing the 8 observed variables above, the results of the second EFA show that the
Kaiser-Meyer-Olkin (KMO) measure is 0.808 > 0.5; Bartlett's Test of Sphericity has a significance of
0.000 < 0.05; the total variance explained is 74.599% > 50%; all Eigenvalues are greater than 1, and
the factor loading coefficients of the observed variables are all greater than 0.4. Ultimately, the 10
observed variables obtained after analysis meet the requirements for convergence and differentiation
are used for correlation and regression analysis. The results of the KMO test, Bartlett's test, and the
factor loading coefficients of the 10 observed variables are presented in the subsequent tables:

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling .808

Adequacy.
Bartlett's Test of Approx. Chi-Square 641.010
Sphericity
df 45

Sig. .000

Table 12: KMO and Bartlett's Test

Rotated Component Matrixa
Component

1 2 3 4
PS1 - I perceive security using my BNPL information .813
through mobile payment platforms.
PS2 - I would feel safe providing sensitive information .842
about myself over the BNPL platforms.
SC1 - I feel that BNPL is very convenient and available .860
at fingertips.
SC2 - I find that BNPL needs to enter personal details .875
only once.

39
 AD1 - I find it hooked to buy unwanted items with .744
BNPL.
AD2 - I tend to buy things that I cannot afford due to .792
BNPL.
AD4 - I do not need to save money because a portion is .788
already paid when purchasing with BNPL.
UF1 - I think the product return policies of BNPL are .849
convenient.
UF2 - I think BNPL has flexible exchange policies. .863
UF3 - I find that BNPL offers a high level of .600
transparency on their public platform.

Table 13: Second Rotated Component Matrixa

After making measurements on dependent variables, the independent ones should also be tested with
EFA, and the results as follows:

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling .811
Adequacy.
Bartlett's Test of Approx. Chi-Square 350.834
Sphericity
df 10

Sig. .000

Table 14: KMO and Bartlett’s Test of Sphericity of Dependent variables

The Kaiser-Meyer-Olkin Measure of Sampling Adequacy (KMO) is 0.811 > 0.5, indicating the
suitability of the dataset for factor analysis. Bartlett's test yields a result of 350.834 with a significance
of 0.000 < 0.05. This outcome suggests a correlation among the observed variables in the population,
affirming the appropriateness of the EFA factor analysis.
Component Matrixa

40
 Component

1
IBT1 - I do not plan in advance what to purchase. .822

IBT2 - I find it unnecessary to consider what to buy when .804

shopping with BNPL.
IBT3 - I usually make purchases based on desire rather than .788
necessity.
IBT4 - I tend to make impulsive purchases without considering .779
the consequences.
IBT5 - I ended up spending more money on my purchases than I .733
intended to.

Table 15: Testing results of EFA of Dependent Variables

The findings from the EFA analysis reveal that using Principal Component Analysis and
Varimax rotation for factor extraction, there are 5 observable variables. The cumulative variance
amounts to 61.759 %, surpassing the 50% threshold. Additionally, the Eigenvalue coefficient stands at
3.088 > 1, and all factor loading coefficients of the observable variables are above 0.3. Consequently,
the scale is deemed satisfactory, and the measurement variables pertaining to " Impulsive Buying
Tendency" will be utilized for subsequent analyses.

4.5. Pearson Correlation: Testing research hypotheses

When examination of the table results at a significance level of 1%, it was found that the significance
values (Sig) were all below 0.05. This indicates that the independent variables have a significant
correlation with the dependent variables.
Correlations

Security Friendliness Convenience Awareness

Pearson .446** .486** .365** .600**

Tendency
Correlation
Sig. (2- .000 .000 .000 .000

41
 tailed)

Table 16: Correlation matrix table

4.6. Multivariate Regression Analysis

4.6.1. Validation of the research model

Conduct multiple regression analysis with a total of 5 adjusted variables, including 5 dependent
variables: Perceived Security (PS); Simple and Convenient Usage (SC); Awareness of BNPL (AD),
User friendliness (UF), Customization Design (CD) and an independent variable: Impulsive Buying
Tendency (IBT), the authors developed a regression model as follows:
IBT =β 0 + β 1∗PS+ β 2∗SC+ β 3∗AD + β 4∗UF + β 5∗CD

Where:

− β 0 is a constant and β i ( i=1 ; 2; 3 ; 4 ) is partial regression coefficient

− IBT: Impulsive Buying Tendency

− PS: Perceived Security

− SC: Simple and Convenient Usage

− AD: Awareness of BNPL

− UF: User friendliness

− CD: Customization Design

However, of the five independent variables above, Customization Design (CD) has no significant
statistical meanings, so the research team decided to perform linear regression for the four variables
left, including: Perceived Security (PS); Simple and Convenient Usage (SC); Awareness of BNPL
(AD), User friendliness (UF) and the independent variable Impulsive Buying Tendency (IBT).

42
 Model Summaryb
Adjusted R Std. Error of Durbin-
Model R R Square Square the Estimate Watson
1 .652a .425 .411 .77020 1.855
a. Predictors: (Constant), PS, SC, AD, UF
b. Dependent Variable: Impulsive Buying Tendency (IBT)

Table 17: Initial Model Summary

In terms of model adequacy, the adjusted R-squared coefficient is 41.1%, as shown in the table
above. This reveals that 41.1% of the fluctuations observed in the dependent variable are influenced by
the independent variables scrutinized in the regression analysis, while the remaining 58.9% can be
attributed to factors beyond those considered in the model, as well as random errors.
To test the relationships in the research model between the dependent and independent
variables, the authors formulate the hypothesis: H 0= β1=β 2=β 3= β 4=β 5=0
ANOVAa
Sum of Mean
Model Squares df Square F Sig.
1 Regression 74.424 4 18.606 31.365 .000b
Residual 100.846 170 .593

Total 175.269 174

a. Dependent Variable: Impulsive Buying Tendency (IBT)

b. Predictors: (Constant), PS, SC, AD, UF
Table 18: ANOVA

The F-test result from the ANOVA analysis, with a significance value of 0.000 (p < 0.05), leads
to the rejection of the null hypothesis (H0). This indicates, with 95% confidence, that at least one
independent variable significantly affects the dependent variable. To determine which specific
independent variables have a statistically significant impact on the outcome, the authors utilize the t-
test. The t-test results presented in Table 28 reveal that three independent variables: Awareness of
BNPL (AD), User Friendliness (UF), and Perceived Security (PS) exhibit statistically significant
effects at the 5% significance level.
Coefficientsa

43
 Unstandardized Standardized Collinearity
Model Coefficients Coefficients t Sig. Statistics
Std.
B Error Beta Tolerance
1 (Constant) .239 .280 .852 .395

Perceived Security .169 .083 .152 2.033 .044 .603

User Friendliness .205 .091 .177 2.250 .026 .546

Simple and .013 .079 .012 .165 .869 .610

Convenient Usage

Awareness of .436 .068 .443 6.456 .000 .717

BNPL

Dependent variable: Impulsive Buying Tendency (IBT)

Table 19: Regression Coefficients

Regarding multicollinearity testing, following the guideline of VIF < 3 (Hair & colleagues, 2010), the
results from the Coefficientsa table all meet this criterion. Thus, there is no multicollinearity issue
observed in the model.

Collinearity
Model Statistics

VIF
1 (Constant)
Perceived Security 1.657

User Friendliness 1.831

Simple and Convenient Usage 1.640

Awareness of BNPL 1.394

44
 Table 20: Collinearity Statistics VIF

From the above results, the Simple and Convenient Usage (SC) variable has little significant meaning
in statistics (0.012 < 0.100), so the author has re-run the linear regression model with only three
variables, they are: Perceived Security (PS); Awareness of BNPL (AD) and User friendliness (UF). The
authors have come up with the correlation between dependent variables and independent variables as in
the following formula:
𝐼𝐵𝑇 = .152∗𝑃𝑆 + .443∗𝐴𝐷 + .177∗𝑈𝐹

Where:

− IBT: Impulsive Buying Tendency

− PS: Perceived Security

− AD: Awareness of BNPL

− UF: User friendliness

Figure 7: The Research Outcome Model

45
 - Normality testing for residuals:
It can be clearly seen in the histogram that the Mean is 2.91E-16 (approximately 0) and the
Standard Deviation is 0.988, which is close to 1. From an overall viewpoint, the residuals form a bell-
shaped distribution, which indicates that the distribution is roughly normal, suggesting that the
assumption of normally distributed residuals holds true.

Figure 8: Normalized Residual Frequency Chart Histogram

Regarding the Normal P-P Plot, if the points align closely along a straight line, it suggests that
the residuals are normally distributed. It is noticeable that most of the residual points are concentrated
closely around the diagonal line. Consequently, it implies that the residuals exhibit an approximate
normal distribution, suggesting that the normal distribution of the residuals are not infringed.

46
 Fi
gure 9: Normalized Residual Frequency Chart Normal P-P Plot

- The linear relationship assumption:

In a regression model, a linear relationship must exist between the independent and dependent
variables. This can be assessed using a Scatter Plot of the standardized residuals versus the predicted
values. If the standardized residuals are randomly distributed around the zero line on the vertical axis,
we can infer that the linear relationship assumption holds true.
Based on the Scatter Plot results, the standardized residuals appear to be randomly distributed
around the zero line, showing no specific pattern. Therefore, we can affirm that the linear relationship
assumption remains valid.

47
 F
igure 10: Normalized Residual Frequency Chart Scatterplot

- Assumption of no correlation among residuals:

The table above indicates a Durbin-Watson value of 1.855. As stated by Hoang Trong & Chu
Nguyen Mong Ngoc (2008), a Durbin-Watson value near 2 suggests that the residuals are not first-
order correlated. Therefore, we can infer that the model's residuals are uncorrelated. Moreover, the
random distribution of standardized residuals in the Scatter Plot further confirms that there is no
correlation among the residuals. Thus, the assumption of no correlation among residuals is not violated.

4.6.2. Research results

Based on the study’s findings, all the hypotheses were confirmed. Therefore, the model reveals
four factors that significantly influence the dependent variables: Perceived Security (PS), Users’
Awareness (UA) and User Friendliness (UF). Among these, Users’ Awareness (UA) has the greatest
impact on the Impulsive Buying Tendencies using BNPL, with a regression coefficient of 0.443. This is
followed by User Friendliness (UF) and Perceived Security (PS), with regression coefficients of 0.177
and 0.152 respectively.

Hypothesis Assumptions Conclusion

48
 Customization Design impacts the Impulsive Buying
H1 Unsupported
Tendency among UEH students

Perceived Security impacts the Impulsive Buying

H2 Supported
Tendency among UEH Students

Simple and Convenient Usage impacts the Impulsive

H3 Unsupported
Buying Tendency among UEH students

Users’ Awareness impacts the Impulsive Buying Supported

H4
Tendency among UEH Students

User Friendliness impacts the Impulsive Buying Supported

H5
Tendency among UEH Students

Table 21: Summary results of hypotheses testing

49
 CHAPTER V: CONCLUSION AND RECOMMENDATION

5.1. Conclusion

This study examines the factors influencing impulsive buying tendencies among UEH students
utilizing BNPL (Buy Now, Pay Later) services. The quantitative phase of the research involved
administering a questionnaire with a 5-point Likert scale to measure one dependent variable and five
independent variables. Following data collection, the research team assessed reliability using
Cronbach's Alpha, conducted Exploratory Factor Analysis, and performed linear regression analysis
with SPSS 25, based on a valid sample size of 175 respondents. The results showed that, after
Cronbach's Alpha reliability test and EFA test, the factor 'Customization design' (CD) was not
statistically significant for the dependent variable. Specifically, all observed variables in this factor
were correlated with other observed variables, so the authors decided to remove this factor. After
further linear regression analysis, the factor 'Simple and convenient usage' (SC) was also not
statistically significant for the dependent variable. Specifically, this factor had a Sig. value of 0.869 > α
= 0.05, so the authors decided to remove this variable from the regression model and conclude that the
null hypothesis H0 for the SC variable was rejected. Therefore, hypothesis H3: Simple and convenient
usage positively impacts on the impulsive buying tendencies of UEH students was rejected, suggesting
that nowadays, the purchasing decisions of UEH students as consumers are not shaped by the
accessibility and user-friendliness of BNPL payment methods. The remaining factors: Perceived
Security, Users' Awareness, and User Friendliness, were all statistically significant and had an impact
on the impulsive buying tendencies of UEH students.

This research report answers questions mentioned in Chapter 1 as follows:

Question 1: Which factors BNPL motivate students to overconsume?

Our report concludes 3 factors that contribute to the phenomenon when consumers make
purchases without any prior planning or consideration are Perceived Security, Users’ Awareness and
User Friendliness. Based on the regression coefficients, the authors concluded that Users' Awareness
had the strongest impact on impulsive buying tendencies of UEH students, followed by User
Friendliness, and finally Perceived Security, which had the weakest impact.

50
Question 2: How do impulsive buying tendencies impact UEH students’ lives?

Purchasing goods based on desire rather than consideration could lead to items not planned in
advance. Students tend to be more prone to indulgence, and impulsive buying can also lead them to
lose their sense of control and dissatisfaction with their current possessions. When students prioritize
impulsive purchases, they could neglect their financial situation and struggle to build good financial
habits like budgeting and saving for the future. This lack of preparation can leave them exposed during
unexpected events or unable to afford long-term goals like education or starting a family.

Question 3: What recommendations do we propose for students to effectively monitor their

spending?

With this question, our group proposed some recommendations in the next part.

5.2. Recommendation

After conducting and receiving the research results, the authors want to provide some
recommendations:
- Most of the individuals who participated in the survey are students from UEH., so the
generalizability of the findings may be limited. Therefore, it is necessary to extend the survey to
students from other universities in HCM city, and to broaden the age spectrum of the
participants in the research. This approach will facilitate the expansion into further related
studies.
- Impulse purchases often lead customers to buy more than they originally intended or needed.
Hence, consumers should become more aware of their purchasing tendency and create healthier
financial habits to avoid the negative impacts of overconsumption.
- While the research showed that perceived security is considered to have a relatively significant
impact, it also suggests the bigger the assumed risk of using BNPL, the more likely consumers
are to avoid adopting the systems. Companies that offer BNPL services should build and
strengthen their BNPL security systems, as well as taking measures to protect consumers. to
foster customers’ greater trust and encourage usages of BNPL services.
- User-friendliness also significantly impacts customers’ impulsive buying tendency when using
BNPL services. As BNPL platforms often offer friendly policies to protect the benefits of
customers, along with transparent public platforms and a variety of promotional initiatives
designed to attract and retain customers. This design and experience of friendly BNPL services

51
 may cause users to be unaware of the actual cost of the purchase, which can lead customers to
underestimate the financial burdens of their impulsive purchases.
- Founded on the research findings, it showed that Users' Awareness of BNPL services have the
most significant negative impact on users’ impulsive purchasing tendency. This implies that as
users become more informed about BNPL, they are less engaging in unplanned purchases. To
raise customers’ awareness, the authors recommend that users should be more aware and learn
more about BNPL practices and policies, features of the various BNPL services available.
Simultaneously, BNPL companies should provide more policies that encourage students to
participate and use a range of BNPL options, providing them with the opportunity to objectively
evaluate, compare, and identify the most suitable option. By understanding these inner works,
customers can actively manage their budget deficits and set a realistic financial plan, result in
the decrease in impulsive buying tendencies and improvement of financial well-being.
- Subsequently, Future studies should investigate deeper into the long-term implications of the
observed impulsive buying tendencies among university students using BNPL services. By
understanding the potential financial consequences, such as debt accumulation, financial stress,
and the impact on credit scores, users can take proactive measures to better manage and
improve their financial well-being.
- Referring to the 17 Sustainable Development Goals established by the United Nations, our
research aims to encourage consumers to be more mindful and responsible in their shopping and
consumption behaviors (SDG 12). By considering their actions carefully, we hope that even
small changes will contribute to stable economic growth (SDG 8) and ensure adequate income
and living conditions for everyone (SDG 1). Our team encourages the following topics to
expand the connection to other sustainable development goals, creating a comprehensive
approach in the research.

5.3. Limitation and Development

Conscious of the time constraint and resources for research, our group has made every effort to
complete the topic seriously and overcome weaknesses in research ability. However, like any other
research, our studies still have some limitations. Therefore, we would like to propose some potential
research directions for future research.
Firstly, the survey sample mainly consists of friends, friends of friends, and relatives, leading to
a concentration of responses from first-year students. Additionally, the survey only focuses on UEH
university students without extending to other age groups. This limits our ability to identify the impact

52
of BNPL on different age groups. The authors suggest conducting the survey using a larger sample size
to achieve results that are more comprehensive and accurately reflect the effects.
Secondly, the scope of this study only focuses on UEH university in Ho Chi Minh City, which
affects the generalizability and comprehensiveness of the research model. Therefore, it is recommended
to conduct surveys on a larger scale at multiple universities in different provinces and cities, especially
in economically developed areas like Hanoi, Da Nang, etc.
Thirdly, the current study only explores familiar factors such as Customization design,
Perceived Security, Simple and convenient usage, Awareness of BNPL, and User friendliness of
BNPL. One noticeable point is that, when running EFA, from 18 observed variables, we eliminated 8
variables, of which 1 was the cause variable, leaving only 10 variables. This is the significant limitation
of the author group. Our group proposes the following research studies should be further explored to
discover observed variables that current research has not addressed, while also considering and
adjusting the scale to achieve better value.
Finally, the study also did not analyze the differences in overspending tendencies using the buy
now, pay later feature by different providers. In this study, the sample only focused on the
characteristics of the feature without categorizing it into which application it belongs to (Momo,
ZaloPay, Shopee Paylater, etc.). It is highly likely that the intention to use each application frequently
will differ. Future research should evaluate which BNPL app has the most impact on consumer
overspending tendencies.

53
 REFERENCES
1. Abdullah, F., Ward, R., & Ahmed, E. (2016). Investigating the influence of the most commonly
used external variables of TAM on students’ Perceived Ease of Use (PEOU) and Perceived
Usefulness (PU) of e-portfolios. Computers in Human Behavior, 63, 75–90.
[Link]
2. Hirschman, E. C. (1979), Differences in Consumer Purchase Behavior by Credit Card Payment
System, Journal of Consumer Research, Volume 6, Issue 1, Pages 58–66,
[Link]
3. Flavián, C. and Guinalíu, M. (2006), "Consumer trust, perceived security and privacy policy:
Three basic elements of loyalty to a web site", Industrial Management & Data Systems, Vol.
106 No. 5, pp. 601-620. [Link]
4. Kotler, P. (1973), “Atmospherics as a marketing tool", Journal of Retailing, Vol. 49 No. 4, pp.
48-64.
[Link]
5. Prelec, D., & Loewenstein, G. (1998). The red and the black: Mental accounting of savings and
debt. Marketing Science, 17 (1), 4–28. [Link]
6. Rook, D. (1987), The Buying Impulse, Journal of Consumer Research, Volume 14, Issue 2,
Pages 189–199, [Link]
7. Rook, D. & Gardner, M. (1993). In the mood: Impulse buying's affective antecedents. Research
in Consumer Behavior 6(7):1-28,
[Link]
e_antecedents
8. Stern, H. (1962). The Significance of Impulse Buying Today. Journal of Marketing, 26(2), 59–
62. [Link]
9. Saleh, M.A. (2012). An Investigation of the Relationship between Unplanned Buying and Post-
purchase Regret. International Journal of Marketing Studies, 4, 106.
10. Thompson, C. S. & Spacapan, S. (1991). Perceptions of Control in Vulnerable Populations.
[Link]
11. Zhang, J.; Luximon, Y.; Song, Y. (2019). The Role of Consumers’ Perceived Security,
Perceived Control, Interface Design Features, and Conscientiousness in Continuous Use of
Mobile Payment Services. Sustainability, 11, 6843. [Link]

54
12. Grabner-Kraeuter, S. (2002). The Role of Consumers’ Trust in Online-Shopping. Journal of
Business Ethics, 39(1/2), 43–50. [Link]
13. Pavlou, P. (2001), "Consumer Intentions to Adopt Electronic Commerce - Incorporating Trust
and Risk in the Technology Acceptance Model". DIGIT 2001 Proceedings.
[Link]
14. Grabner-Kraeuter, S. The Role of Consumers' Trust in Online-Shopping. Journal of Business
Ethics 39, 43–50 (2002). [Link]
15. Thompson, C. (2023), Understanding the dangers of using Buy Now, Pay Later this Christmas,
PayPlan, blog, [Link]
pay-later-this-christmas/
16. Boyington, A. (2024), Pros & Cons of ‘Buy Now, Pay Later’, HomeOwner Website.
[Link]
17. Hartman, R. (2021). "Buy Now, Pay Later: How BNPL is Transforming Online Shopping",
BigCommerce, [Link]
18. Salisbury, W.D., Pearson, R.A., Pearson, A.W. and Miller, D.W. (2001), "Perceived security
and World Wide Web purchase intention", Industrial Management & Data Systems, Vol. 101
No. 4, pp. 165-177. [Link]
19. Agustin, L. (2022). Evaluation of Factors influencing the Shopee PayLater adoption in
Indonesia. Management Science and Business Decisions, 2(2), 5–18.
[Link]
20. Hartono, E.; Holsapple, C.W.; Kim, K.Y.; Na, K.S.; Simpson, J.T. Measuring perceived
security in B2C electronic commerce website usage: A respecification and validation. Decis.
Support Syst. 2014, 62, 11–21. [Link]
21. Marathe, S. & Sundar, S. (2011). What drives customization?: control or identity?, CHI '11:
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, Pages 781–
790. [Link]
22. Fisher, C., Holland, C., & West, T. (2021). Developments in the buy now, pay later market.
Reserve Bank of Australia Bulletin, pp. 1-17.
23. Nindya, I., & Astuti, R. D. (2022). Factors affecting intention to use of online installment
payment Paylater. The 6th International Conference on Family Business and Entrepreneurship,
ICFBE 2022, pp. 1-12.

55
24. O’Brien, L., Ramsay, I., & Ali, P. (2022). The Role of Credit Cards, Payday Loans, Consumer
Leases and Buy-Now-Pay-Later Products in Personal Insolvency. Insolvency Law Journal, Vol.
30, No. 3, pp. 160-170.
25. Ah Fook, L. and McNeill, L. (2020a) ‘Click to buy: The impact of retail credit on over-
consumption in the online environment’, Sustainability, 12(18), p. 7322.
26. Relja, Ruffin & Ward, Philippa & Zhao, Anita. (2023). Understanding the psychological
determinants of buy-now-pay-later (BNPL) in the UK. International Journal of Bank Marketing.
42. pp. 10.
27. Vijay Amrit Raj, Sahil Singh Jasrotia & Siddharth Shankar Rai. (2023) Intensifying materialism
through buy-now pay-later (BNPL): examining the dark sides. International Journal of Bank
Marketing.
28. Verplanken, B. and Herabadi, A. (2001) ‘Individual differences in impulse buying tendency:
Feeling and no thinking’, European Journal of Personality, 15(1_suppl).
29. Gourville, J. T., & Soman, D. (1998). Payment depreciation: The behavioral effects of
temporally separating payments from consumption. Journal of Consumer Research, 25(2), 160–
174.
30. Falk, Armin & Becker, Anke & Dohmen, Thomas & Enke, Benjamin & Huffman, David &
Sunde, Uwe. (2018). Global Evidence on Economic Preferences*. Quarterly Journal of
Economics. 133. 1645-1692.
31. Kim, Changsu & Galliers, Robert & Shin, Namchul & Ryoo, Joohan & Kim, Jongheon. (2012).
Factors influencing Internet shopping value and customer repurchase intention. Electronic
Commerce Research and Applications. 11. 374–387
32. Edmunds, S. (2018) ‘Houses of contention:tar babyandessence’, American Literature, 90(3),
pp. 613–641.
33. Sweeney, L. (2024) What is buy now, pay later?, Forbes. Available at:
[Link] (Accessed: 23 May 2024).
34. Lake, R. (2023) Buy now, pay later (BNPL): What it is, how it works, Pros and Cons,
Investopedia. Available at: [Link]
(Accessed: 23 May 2024).
35. Streaks, J. (2024) 3 reasons why a financial planner advises against using buy now, pay later,
even though it’s practically everywhere, Business Insider. Available at:
[Link]
2024-1 (Accessed: 23 May 2024).

56
36. Altman N. (2023) Buy now, pay later services: August 2022 Nationally Representative Internet
Survey, CR Advocacy. Available at: [Link]
pay-later-services-august-2022-nationally-representative-internet-survey/ (Accessed: 23 May
2024).
37. Hoàng Trọng, Chu Nguyễn Mộng Ngọc (2008), Phân tích dữ liệu nghiên cứu với SPSS, NXB
Hồng Đức 2008
38. Tabachnick, B. G., & Fidell, L. S. (1996). Using Multivariate Statistics (3rd ed.). New York:
Harper Collins.
39. Field, A. (2009) Discovering Statistics Using SPSS. 3rd Edition, Sage Publications Ltd.,
London.
40. Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2009). Multivariate data analysis.
Pearson Prentice Hall.
41. Anh et al (2023). Các nhân tố ảnh hưởng đến ý định sử dụng dịch vụ mua trước trả sau trên các
sàn thương mại điện tử của sinh viên trên địa bàn Hà Nội. (Page 7)
[Link]
42. Jiaxin Zhang, Yan Luximon and Yao Song (2019). The Role of Consumers’ Perceived Security,
Perceived Control, Interface Design Features, and Conscientiousness in Continuous Use of
Mobile Payment Services, 4-7.
[Link]
43. Goonetilleke U. (2022) “BUY NOW PAY LATER” Services and the financial wellbeing of
young Australian adults, 70-77.
[Link]
44. Talaat Abo Elmaaty Ibrahim Ahmed (2023). Antecedents and consequences of impulsive
buying behavior (a field study), 51-52. [Link]
45. Dobson, A., B. Robards, and N. Carah. 2018. Digital Intimate Publics and Social Media.
46. Muchammad A. Z., Lantara I. W. N., [Link]., Ph.D., [Link]., Ph.D., (2023). The Effect of Buy
Now Pay Later on The Overspending and Overconsumption of Users in E-commerce Settings
[Link]
47. Della Ayu Zonna Lia, Salsabilla Lu’ay Natswa, (2021). Buy-Now-Pay-Later (BNPL):
Generation Z’s Dilemma on Impulsive Buying and Overconsumption Intention
[Link]
48. Chay Fisher, Cara Holland, and Tim West, (2021). Developments in the Buy Now, Pay Later
Market [Link]
[Link]

57
49. King, A. (2021). The impact of Buy Now, Pay Later on consumer purchasing behavior. Journal
of Consumer Psychology, 45(3), 452-466.

58
 APPENDIX
Appendix 1: Research Questionnaire

“Chào bạn, chúng mình là sinh viên khóa 49 của Đại học Kinh tế TP. HCM (UEH), thuộc chuyên
ngành Kiểm Toán. Hiện tại, nhóm chúng mình đang thực hiện nghiên cứu khảo sát về tính năng "Buy
now - Pay later" của sinh viên UEH để phục vụ cho bài dự án nghiên cứu khoa học

Chúng mình mong nhận được sự ủng hộ và một chút thời gian đóng góp của các bạn để hoàn thành
việc khảo sát này!! Chúng mình cam kết 100% thông tin thu thập được sẽ chỉ được sử dụng cho mục
đích nghiên cứu mang tính giáo dục, học thuật và sẽ được bảo mật trong suốt quá trình nghiên cứu.

Nhóm xin chân thành cảm ơn bạn!!”

Phần 1: Thông tin cá nhân

Giới tính của bạn là:

o Nam
o Nữ

Bạn là sinh viên năm mấy?

o Năm 1
o Năm 2
o Năm 3
o Đã tốt nghiệp

Bạn có biết đến chức năng "Buy Now Pay Later" không?

o Có
o Không

Nếu có, bạn đã bao giờ sử dụng chức năng BNPL để thanh toán/mua hàng chưa?

o Có
o Không

59
Phần 2: Khảo sát về các yếu tố liên quan đến mua hàng ngẫu hứng khi sử dụng dịch vụ “Buy
Now Pay Later”
Với các phát biểu dưới đây, các bạn vui lòng bày tỏ quan điểm theo thang đo từ 1-5 (từ “Rất không
đồng ý” đến “Rất đồng ý”)
Tuỳ chỉnh giao diện

1 2 3 4 5

Tôi cảm thấy rằng tôi có thể

tùy chọn các phương thức
thanh toán theo mong muốn
của mình.

Tôi nhận thấy rằng tôi có thể

đảm bảo sự riêng tư theo
mong muốn mình

Tôi cảm thấy rằng tôi có thể

cài đặt bảo mật theo mong
muốn mình

Mức độ an ninh

1 2 3 4 5

Tôi nhận thức được sự bảo

mật khi sử dụng thông tin
BNPL của mình thông qua
nền tảng thanh toán di động.

Tôi sẽ cảm thấy an toàn khi

cung cấp thông tin nhạy cảm
về bản thân trên nền tảng
BNPL.

Tôi hiểu rằng nền tảng BNPL

60
 là hệ thống an toàn để thực
hiện giao dịch.

Đơn giản và tiện dụng

1 2 3 4 5

Tôi cảm thấy BNPL rất tiện

lợi và có sẵn trong tầm tay

Tôi thấy rằng BNPL chỉ cần

nhập thông tin cá nhân

một lần

Tôi thấy rằng BNPL có dịch

vụ khách hàng tuyệt vời mang
lại cảm giác thân thiện

Tôi thấy BNPL có hệ thống

rất rõ ràng, minh bạch với
những lời nhắc nhở, tuyên bố

thường xuyên

Sự nhận thức về “Mua Trước Trả Sau”

1 2 3 4 5

Tôi cảm thấy BNPL rất tiện

lợi và có sẵn trong tầ Tôi cảm
thấy cuốn hút khi mua những
món không cần thiết khi sử
dụng BNPL

Tôi có xu hướng mua những

thứ mà tôi không đủ khả năng

61
 chi trả do BNPL

Tôi thường không quan tâm

đến sự thúc đẩy tiêu dùng của
tính năng BNPL

Tôi cảm thấy không cần tiết

kiệm lại sau khi chi một
khoản tiền để mua sắm bắng
BNPL

Sự thân thiện với người dùng

1 2 3 4 5

Tôi thấy chính sách trả sản

phẩm của BNPL rất thuận
tiện.

Tôi nghĩ BNPL có chính sách

đổi sản phẩm linh hoạt.

Khi sử dụng BNPL, tôi thấy

rằng BNPL cung cấp sự minh
bạch hoàn toàn về các khoản
trả góp trong tương lai của
bạn ngay từ đầu.

Tôi thấy rằng BNPL cung cấp

mức độ minh bạch cao trên
nền tảng công khai của họ.

62
Phần 3: Khảo sát về xu hướng mua hàng mất kiểm soát - hậu quả của việc mua hàng ngẫu hứng
khi sử dụng dịch vụ “Bay Now Pay Later”
Với các phát biểu dưới đây, các bạn vui lòng bày tỏ quan điểm theo thang đo từ 1-5 (“Rất không đồng
ý” đến “Rất đồng ý”)

1 2 3 4 5

Tôi không lên kế hoạch trước

những gì sẽ mua.

Tôi thấy không cần thiết phải

cân nhắc nên mua gì khi dùng
BNPL.

Tôi thường mua hàng dựa trên

sở thích hơn là độ hữu dụng

Tôi có xu hướng mua sắm bốc

đồng mà không nghĩ đến hậu
quả

Sau cùng, tôi đã chi nhiều tiền

hơn cho việc mua sắm so với
dự định ban đầu

63
Appendix 2: Descriptive Statistical Results

Frequency of people knowing BNPL features:

“Bạn có biết đến chức năng "Buy Now Pay Later" không?”
Cumulative
Frequency Percent Valid Percent Percent
Valid Có 205 88.0 88.0 88.0
Không 28 12.0 12.0 100.0

Total 233 100.0 100.0

Frequency of people having used BNPL:

“Nếu có, bạn đã bao giờ sử dụng chức năng BNPL để thanh toán/mua hàng
chưa?”
Cumulative
Frequency Percent Valid Percent Percent
Valid Có 175 85.4 85.4 85.4
Không 30 14.6 14.6 100.0

Total 205 100.0 100.0

Frequency of gender participated in the survey:

Giới tính
Cumulative
Frequency Percent Valid Percent Percent
Valid Nam 98 56.0 56.0 56.0
Nữ 77 44.0 44.0 100.0

Total 175 100.0 100.0

64
Frequency of student level in university:

“Bạn là sinh viên năm mấy:”

Cumulative
Frequency Percent Valid Percent Percent
Valid Năm 1 97 55.4 55.4 55.4
Năm 2 47 26.9 26.9 82.3

Năm 3 31 17.7 17.7 100.0

Total 175 100.0 100.0

Descriptive statistics of the observed variables:

Descriptive Statistics

N Minim Maxim Mean Std. Variance

um um Deviation

Tôi cảm thấy rằng tôi 175 1 5 3.73 1.142 1.304

có thể tùy chọn các
PTTT theo mong
muốn.

Tôi nhận thấy rằng tôi 175 1 5 3.37 1.238 1.532

có thể đảm bảo sự
riêng tư theo mong
muốn.

Tôi cảm thấy rằng tôi 175 1 5 3.62 1.101 1.213

có thể cài đặt bảo mật
theo mong muốn.

Tôi nhận thức được sự 175 1 5 3.47 1.087 1.181

bảo mật khi sử dụng
TT BNPL của mình
thông qua nền tảng

65
TTDĐ.

Tôi sẽ cảm thấy an toàn 175 1 5 3.10 1.153 1.330

khi cung cấp TT nhạy
cảm về bản thân trên
nền tảng BNPL.

Tôi hiểu rằng nền tảng 175 1 5 3.47 1.071 1.147

BNPL là hệ thống an
toàn để thực hiện GD.

Tôi cảm thấy BNPL rất 175 1 5 3.72 1.153 1.329

tiện lợi và có sẵn trong
tầm tay

Tôi thấy rằng BNPL 175 1 5 3.61 1.198 1.434

chỉ cần nhập TT cá
nhân một lần

Tôi thấy rằng BNPL có 175 1 5 3.42 1.116 1.245

DVKH tuyệt vời mang
lại cảm giác thân thiện

Tôi thấy BNPL có hệ 175 1 5 3.59 1.115 1.242

thống rất rõ ràng, minh
bạch với những lời
nhắc nhở, tuyên bố
thường xuyên

Tôi cảm thấy cuốn hút 175 1 5 3.13 1.208 1.460

khi mua những món
không cần thiết khi sử
dụng BNPL

Tôi có xu hướng mua 175 1 5 3.05 1.261 1.589

những thứ mà tôi
không đủ khả năng chi

66
trả do BNPL

Tôi thường không quan 175 1 5 3.34 1.216 1.479

tâm đến sự thúc đẩy
tiêu dùng của tính năng
BNPL

Tôi cảm thấy không 175 1 5 3.07 1.276 1.627

cần tiết kiệm lại sau khi
chi một khoản tiền để
mua sắm bắng BNPL

Tôi thấy chính sách trả 175 1 5 3.57 1.101 1.212

sản phẩm của BNPL rất
thuận tiện.

Tôi nghĩ BNPL có CS 175 1 5 3.52 1.093 1.194

đổi sản phẩm linh hoạt.

Khi sử dụng BNPL, tôi 175 1 5 3.59 1.136 1.290

thấy rằng BNPL cung
cấp sự minh bạch hoàn
toàn về các khoản trả
góp trong tương lai
ngay từ đầu.

Tôi thấy rằng BNPL 175 1 5 3.67 1.111 1.234

cung cấp mức độ minh
bạch cao trên nền tảng
công khai của họ.

Tôi không lên KH 175 1 5 2.95 1.332 1.773

trước những gì sẽ mua.

Tôi thấy không cần 175 1 5 2.58 1.265 1.601

thiết phải cân nhắc nên
mua gì khi dùng

67
 BNPL.

Tôi thường mua hàng 175 1 5 2.95 1.224 1.497

dựa trên sở thích hơn là
độ hữu dụng

Tôi có xu hướng mua 175 1 5 2.86 1.290 1.663

sắm bốc đồng mà
không nghĩ đến hậu
quả

Tôi đã chi nhiều tiền 175 1 5 3.31 1.277 1.631

hơn cho việc mua sắm
so với dự định ban đầu

Valid N (listwise) 175

Appendix 3: Scale reliability

Independent variable 1: “Customization Design”

Case Processing Summary

N %
Cases Valid 175 100.0

Excludeda 0 .0

Total 175 100.0

a. Listwise deletion based on all variables in the procedure.

Reliability Statistics

Cronbach's Alpha N of Items

68
 .799 3

Item-Total Statistics
Scale Corrected Cronbach's
Scale Mean if Variance if Item-Total Alpha if Item
Item Deleted Item Deleted Correlation Deleted
CD1 - Tôi cảm thấy rằng 6.99 4.345 .634 .736
tôi có thể tùy chọn các
PTTT theo mong muốn.
CD2 - Tôi nhận thấy 7.35 3.792 .693 .672
rằng tôi có thể đảm bảo
sự riêng tư theo mong
muốn.
CD 3 - Tôi cảm thấy 7.09 4.578 .610 .761
rằng tôi có thể cài đặt
bảo mật theo mong
muốn.

Independent variable 2: “Perceived Security”

Case Processing Summary

N %
Cases Valid 175 100.0
Excludeda 0 .0

Total 175 100.0

a. Listwise deletion based on all variables in the procedure.

Reliability Statistics

Cronbach's Alpha N of Items

69
 .756 3

Item-Total Statistics
Scale Corrected Cronbach's
Scale Mean if Variance if Item-Total Alpha if Item
Item Deleted Item Deleted Correlation Deleted
PS1 - Tôi nhận thức 6.57 3.775 .572 .688
được sự bảo mật khi sử
dụng TT BNPL của
mình thông qua nền tảng
TTDĐ.
PS2 - Tôi sẽ cảm thấy an 6.94 3.353 .637 .611
toàn khi cung cấp TT
nhạy cảm về bản thân
trên nền tảng BNPL.
PS3 - Tôi hiểu rằng nền 6.57 3.902 .549 .713
tảng BNPL là hệ thống
an toàn để thực hiện GD.

Independent variable 3: “Simple and Convenient Usage”

Case Processing Summary

N %
Cases Valid 175 100.0

Excludeda 0 .0

Total 175 100.0

a. Listwise deletion based on all variables in the procedure.

70
 Reliability Statistics

Cronbach's Alpha N of Items

.720 4

Item-Total Statistics
Scale Corrected Cronbach's
Scale Mean if Variance if Item-Total Alpha if Item
Item Deleted Item Deleted Correlation Deleted
SC1 - Tôi cảm thấy 10.62 6.489 .614 .594
BNPL rất tiện lợi và có
sẵn trong tầm tay
SC2 - Tôi thấy rằng 10.73 6.703 .530 .646
BNPL chỉ cần nhập TT
cá nhân một lần
SC3 - Tôi thấy rằng 10.93 6.793 .582 .615
BNPL có DVKH tuyệt
vời mang lại cảm giác
thân thiện
SC4 - Tôi thấy BNPL có 10.75 8.109 .326 .759
hệ thống rất rõ ràng,
minh bạch với những lời
nhắc nhở, tuyên bố
thường xuyên

Independent variable 4: “Awareness of BNPL”

Case Processing Summary

N %
Cases Valid 175 100.0
Excludeda 0 .0

Total 175 100.0

a. Listwise deletion based on all variables in the procedure.

Reliability Statistics

71
 Cronbach's
Alpha N of Items
.724 4

Item-Total Statistics
Scale Corrected Cronbach's
Scale Mean if Variance if Item-Total Alpha if Item
Item Deleted Item Deleted Correlation Deleted
AD1 - Tôi cảm thấy 9.46 8.158 .557 .637
cuốn hút khi mua những
món không cần thiết khi
sử dụng BNPL
AD2 - Tôi có xu hướng 9.54 7.629 .609 .602
mua những thứ mà tôi
không đủ khả năng chi
trả do BNPL
AD3 - Tôi thường không 9.25 9.373 .350 .752
quan tâm đến sự thúc
đẩy tiêu dùng của tính
năng BNPL
AD4 - Tôi cảm thấy 9.53 7.917 .545 .642
không cần tiết kiệm lại
sau khi chi một khoản
tiền để mua sắm bắng
BNPL

72
Independent variable 5: “User Friendliness”

Case Processing Summary

N %
Cases Valid 175 100.0

Excludeda 0 .0

Total 175 100.0

a. Listwise deletion based on all variables in the procedure.

Reliability Statistics

Cronbach's Alpha N of Items

.788 4

Item-Total Statistics
Scale Corrected Cronbach's
Scale Mean if Variance if Item-Total Alpha if Item
Item Deleted Item Deleted Correlation Deleted
UF1 - Tôi thấy chính 10.78 7.163 .623 .721
sách trả sản phẩm của
BNPL rất thuận tiện.
UF2 - Tôi nghĩ BNPL có 10.83 7.315 .598 .734
CS đổi sản phẩm linh
hoạt.
UF3 - Khi sử dụng 10.76 6.781 .672 .695
BNPL, tôi thấy rằng
BNPL cung cấp sự minh
bạch hoàn toàn về các
khoản trả góp trong

73
 tương lai ngay từ đầu.
UF4 - Tôi thấy rằng 10.68 7.759 .493 .786
BNPL cung cấp mức độ
minh bạch cao trên nền
tảng công khai của họ.

Dependent variable: “Impulsive Buying Tendencies”

Reliability Statistics

Cronbach's Alpha N of Items

.845 5

Item-Total Statistics
Scale Corrected Cronbach's
Scale Mean if Variance if Item-Total Alpha if Item
Item Deleted Item Deleted Correlation Deleted
IBT1 - Tôi không lên kế 11.70 16.003 .695 .801
hoạch trước những gì sẽ
mua.
IBT2 - Tôi thấy không 12.07 16.650 .671 .808
cần thiết phải cân nhắc
nên mua gì khi dùng
BNPL.
IBT3 - Tôi thường mua 11.70 17.072 .654 .813
hàng dựa trên sở thích
hơn là độ hữu dụng
IBT4 - Tôi có xu hướng 11.79 16.705 .646 .814
mua sắm bốc đồng mà
không nghĩ đến hậu quả
IBT5 - Tôi đã chi nhiều 11.34 17.282 .591 .829
tiền hơn cho việc mua

74
 sắm so với dự định.

Appendix 4: Exploratory Factor Analysis (EFA)

EFA of Independent and Dependent Variables for the first time

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling

.862
Adequacy.

Bartlett's Test of
Approx. Chi-Square 1431.749
Sphericity

df 153

Sig. .000

Total Variance Explained

Component Initial Eigenvalues Extraction Sums of Squared Loadings
% of Cumulative % of Cumulative
Total Variance % Total Variance %
1 6.961 38.673 38.673 6.961 38.673 38.673
2 1.496 8.309 46.982 1.496 8.309 46.982
3 1.368 7.600 54.582 1.368 7.600 54.582
4 1.113 6.185 60.767 1.113 6.185 60.767
5 .959 5.330 66.098 .959 5.330 66.098
6 .924 5.132 71.230
7 .798 4.433 75.663
8 .734 4.076 79.739
9 .553 3.070 82.809
10 .530 2.944 85.753
11 .472 2.624 88.377
12 .421 2.337 90.714

75
13 .392 2.179 92.893
14 .325 1.806 94.699
15 .266 1.478 96.178
16 .257 1.430 97.607
17 .243 1.349 98.957
18 .188 1.043 100.000

Rotated Component Matrixa

Component

1 2 3 4 5

CD1 - Tôi có thể tùy .419 .429 .484

chọn các PTTT theo
mong muốn.

CD2 - Tôi có thể đảm .623

bảo sự riêng tư theo
mong muốn.

CD3 - Tôi có thể cài đặt .645 .486

bảo mật theo mong
muốn.

PS1 - Tôi nhận thức .685

được sự bảo mật khi sử
dụng thông tin BNPL
của mình thông qua
TTDĐ.

PS2 - Tôi sẽ cảm thấy an .757

toàn khi cung cấp TT
nhạy cảm về bản thân
trên nền tảng BNPL.

76
PS3 - Tôi hiểu rằng nền .514 .584
tảng BNPL là hệ thống
an toàn để thực hiện GD.

SC1 - Tôi cảm thấy .798

BNPL rất tiện lợi và có
sẵn trong tầm tay

SC2 - Tôi thấy rằng .837

BNPL chỉ cần nhập TT
cá nhân một lần

SC3 - Tôi thấy rằng .466

BNPL có DVKH tuyệt
vời mang lại cảm giác
thân thiện

SC4 - Tôi thấy BNPL có .457 .573

hệ thống rất rõ ràng,
minh bạch

AD1 - Tôi thấy cuốn hút .648

khi mua những món
không cần thiết khi sử
dụng BNPL

AD2 - Tôi thường mua .727

những thứ mà tôi không
đủ khả năng chi trả do
BNPL

AD3 - Tôi không quan .743

tâm đến sự thúc đẩy tiêu
dùng của tính năng
BNPL

77
 AD4 - Tôi cảm thấy .821
không cần tiết kiệm lại
sau khi mua sắm bằng
BNPL

UF1 - Tôi thấy CS trả .754

sản phẩm của BNPL rất
thuận tiện.

UF2 - Tôi nghĩ BNPL có .808

CS đổi sản phẩm linh
hoạt.

UF3 - Tôi thấy rằng .601

BNPL cung cấp sự minh
bạch về các khoản trả
góp.

UF4 - Tôi thấy rằng .588

BNPL cung cấp mức độ
minh bạch cao.

Exploratory Factor Analysis after excluding variables (second linear regression test)

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling .808

Adequacy.

Bartlett's Test of Approx. Chi-Square 641.010

Sphericity

df 45

Sig. .000

78
79
 Total Variance Explained

Comp Initial Eigenvalues Extraction Sums of Squared Loadings

onent

Total % of Cumulative Total % of Cumulative

Variance % Variance %

1 4.210 42.098 42.098 4.210 42.098 42.098

2 1.213 12.133 54.231 1.213 12.133 54.231

3 1.068 10.684 64.916 1.068 10.684 64.916

4 .968 9.683 74.599 .968 9.683 74.599

5 .611 6.108 80.707

6 .540 5.397 86.104

7 .461 4.608 90.711

8 .380 3.802 94.513

9 .281 2.810 97.323

10 .268 2.677 100.000

Rotated Component Matrixa

Component

1 2 3 4

PS1 - Tôi nhận thức được sự bảo .813

mật khi sử dụng thông tin BNPL
của mình thông qua TTDĐ.

PS2 - Tôi sẽ cảm thấy an toàn khi .842

cung cấp TT nhạy cảm về bản
thân trên nền tảng BNPL.

80
SC1 - Tôi cảm thấy BNPL rất .860
tiện lợi và có sẵn trong tầm tay

SC2 - Tôi thấy rằng BNPL chỉ .875

cần nhập TT cá nhân một lần

AD1 - Tôi thấy cuốn hút khi mua .744

những món không cần thiết khi
sử dụng BNPL

AD2 - Tôi thường mua những thứ .792

mà tôi không đủ khả năng chi trả
do BNPL

AD4 - Tôi cảm thấy không cần .788

tiết kiệm lại sau khi mua sắm
bằng BNPL

UF1 - Tôi thấy CS trả sản phẩm .849

của BNPL rất thuận tiện.

UF2 - Tôi nghĩ BNPL có CS đổi .863

sản phẩm linh hoạt.

UF3 - Tôi thấy rằng BNPL cung .600

cấp sự minh bạch về các khoản
trả góp.

Appendix 5: Correlation and Regression Analysis

Correlations

81
 Xu hướng An Thân Tiện Nhận
(biến kq) ninh thiện dụng 2 thức 2

Xu hướng Pearson 1 .446** .486** .365** .600**

(biến kq) Correlation

Sig. (2-tailed) .000 .000 .000 .000

N 175 175 175 175 175

An ninh Pearson .446** 1 .554** .531** .427**

Correlation

Sig. (2-tailed) .000 .000 .000 .000

N 175 175 175 175 175

Thân thiện Pearson .486** .554** 1 .561** .490**

Correlation

Sig. (2-tailed) .000 .000 .000 .000

N 175 175 175 175 175

Tiện dụng Pearson .365** .531** .561** 1 .390**

Correlation

Sig. (2-tailed) .000 .000 .000 .000

N 175 175 175 175 175

Nhận thức Pearson .600** .427** .490** .390** 1

Correlation

Sig. (2-tailed) .000 .000 .000 .000

N 175 175 175 175 175

**. Correlation is significant at the 0.01 level (2-tailed)

Appendix 6: Linear Regression Analysis

82
 Model Summaryb

Model R R Square Adjusted R Std. Error of Durbin-

Square the Estimate Watson

1 .652a .425 .411 .77020 1.855

a. Predictors: (Constant), Nhận thức, Tiện dụng, An ninh, Thân thiện

b. Dependent Variable: Xu hướng (biến kết quả)

ANOVAa
Sum of Mean
Model Squares df Square F Sig.
1 Regression 74.424 4 18.606 31.365 .000b
Residual 100.846 170 .593

Total 175.269 174

a. Dependent Variable: Xu hướng (biến kết quả)

b. Predictors: (Constant), Nhận thức, Tiện dụng, An ninh, Thân thiện

Model Unstandardized Standardize t Sig. Collinearit Collinearity

Coefficients d y Statistics Statistics
Coefficients

B Std. Beta Tolerance VIF

Error

1 Constant .239 .280 .852 .395

An ninh .169 .083 .152 2.033 .044 .603 1.657

Thân .205 .091 .177 2.250 .026 .546 1.831

thiện

83
 Tiện dụng .013 .079 .012 .165 .869 .610 1.640
2

Nhận .436 .068 .443 6.456 .000 .717 1.394

thức 2

a. Dependent Variable: Xu hướng (biến kết quả)

Normalized Residual Frequency Chart Histogram

Normalized Residual Frequency Chart Normal P-P Plot

84
Normalized Residual Frequency Chart Scatterplot

85