ĐẠI HỌC KINH TẾ TP.

HỒ CHÍ MINH
KHOA KINH DOANH QUỐC TẾ - MARKETING

BÁO CÁO TỔNG KẾT

ĐỀ TÀI GIẢI THƯỞNG THỰC HÀNH NGHIÊN
CỨU KHOA HỌC SINH VIÊN KHOA KINH DOANH
QUỐC TẾ - MARKETING NĂM 2024

< FACTORS AFFECTING THE DISHONEST

BEHAVIOR OF UNIVERSITY STUDENTS IN
HO CHI MINH CITY >

Thuộc nhóm: : Bùi Thu An - 31221021644

Thái Ngọc Bảo - 31221026573
Huỳnh Quốc Huy - 31221023712
Nguyễn Hùng Long - 31221020421
Trần Thanh Thảo - 31221021683

MÃ SỐ ĐỀ TÀI (do Ban Tổ chức ghi):……………..

TP. Hồ Chí Minh, 29/08/2023

 I

TABLE OF CONTENTS

LIST OF ABBREVIATIONS ................................................................................................. III

LIST OF FIGURES ................................................................................................................IV
LIST OF TABLES .................................................................................................................... V
ABSTRACT ................................................................................................................................ 1
CHAPTER 1 - INTRODUCTION............................................................................................. 2
1.1 Reasons for choosing the topic ................................................................................................... 2
1.1.1 Overview of the problem ........................................................................................................................2
1.1.2 Necessity of the research ........................................................................................................................5

1.2 Research gap ................................................................................................................................ 6

1.3 Research questions ...................................................................................................................... 7
1.4 Research objectives...................................................................................................................... 7
1.5 Research subjects & scope .......................................................................................................... 8
1.6 Research methodology ................................................................................................................ 8
1.7 Research structure ....................................................................................................................... 8
1.8 Research contribution ................................................................................................................. 9
1.8.1 Theoretical contributions .......................................................................................................................9
1.8.2 Practical contributions ...........................................................................................................................9

CHAPTER 2 - LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT ........... 11

2.1 Relevant research ...................................................................................................................... 11
2.2 Key concepts ............................................................................................................................... 13
2.2.1 Term Definitions: .................................................................................................................................13
2.2.2 Variable Definitions .............................................................................................................................14
2.3 Theoretical background ............................................................................................................ 16
2.3.1 The theory of planned behavior ...........................................................................................................16
2.3.2 Horizontal and Vertical Dimensions of Individualism and Collectivism .............................................18

2.4 Hypothesis development............................................................................................................ 19

2.5 Proposed research model .......................................................................................................... 25
CHAPTER 03 - RESEARCH METHODOLOGY .................................................................. 26
3.1. Procedure .................................................................................................................................. 26
3.2. Qualitative Research ................................................................................................................ 26
3.3. Quantitative Research .............................................................................................................. 27
3.3.1. Research Subject .................................................................................................................................27
3.3.2. Sample Size ..........................................................................................................................................27
3.3.3. Sample Selection and Data Collection Method...................................................................................27
3.3.4. Data Analysis Method .........................................................................................................................28
 II
3.4. Measurement scale ................................................................................................................... 29
3.5. Data Analysis Process ............................................................................................................... 35
3.5.1. Descriptive statistics analysis .............................................................................................................35
3.5.2. Measurement Model ............................................................................................................................36
3.5.3. Assess the quality of observed variables .............................................................................................36
3.5.4. Assessing Cronbach’s Alpha coefficient .............................................................................................36
3.5.5. Assessing Composite Reliability..........................................................................................................37
3.5.6. Assessing Convergent validity .............................................................................................................38
3.5.7. Assessing Discriminant Validity..........................................................................................................38

3.6. Assessing Structural Model ..................................................................................................... 39

3.6.1. Assessing Multicollinearity .................................................................................................................39
3.6.2. Relationship in structural model .........................................................................................................40
3.6.3. Assessing Coefficient of determination (R2) ........................................................................................40
3.6.4. Assessing Effect Size (f2)......................................................................................................................41

CHAPTER 4 - DATA ANALYSIS, RESULTS, AND DISCUSSIONS ................................. 42

4.1 Sample characteristics ....................................................................................................... 42
4.2 Descriptive statistics .................................................................................................................. 44
4.3 Reliability and discriminant validity ....................................................................................... 45
4.4 Common method bias ................................................................................................................ 50
4.5 Assessing Structural Model ...................................................................................................... 51
4.5.1 Hypothesis testing ................................................................................................................................51
4.5.2 Moderating variables ...........................................................................................................................52
4.5.3 Quality testing ......................................................................................................................................54

4.6 Discussion ................................................................................................................................... 56

CHAPTER 5 – CONCLUSIONS AND IMLICATIONS ....................................................... 64
5.1 Discussion ................................................................................................................................... 64
5.2 Research contributions.............................................................................................................. 65
5.2.1 Theoretical contributions .....................................................................................................................65
5.2.2 Pratical implications ............................................................................................................................65
5.3 Limitations and Further Research ........................................................................................... 66
REFERENCES ........................................................................................................................ 68
 III
LIST OF ABBREVIATIONS

Notation Full Definition

AI Artificial Intelligence
AT Attitude
HVIC Horizontal-Vertical Individualism-Collectivism
IT Intention
JT Justification
MO Moral Obligation
PBC Perceived behavioral control
SN Subjective norms
TPB Theory of Planned Behavior
 IV
LIST OF FIGURES

Figure 1. Research model of Stone et al. (2009) ......................................................... 11

Figure 2. Research model of Matthew et al. (2009) ................................................... 12
Figure 3. Research model of Yinxia (2023) ................................................................ 13
Figure 4. Proposed research model .............................................................................. 25
Figure 5. The PLS-SEM Results.................................................................................. 54
 V
LIST OF TABLES

Table 1. Summary of research hypotheses................................................................... 24

Table 2. Questionnaire ................................................................................................. 30
Table 3. Gender ............................................................................................................ 42
Table 4. University and College .................................................................................. 42
Table 5. Major .............................................................................................................. 43
Table 6. Academic year ............................................................................................... 43
Table 7. Descriptive statistics ...................................................................................... 44
Table 8. Outer loadings ................................................................................................ 46
Table 9. Construct Reliability and Validity ................................................................. 48
Table 10. Correlations between Research Constructs (Fornell - Larcker Criterion) ... 48
Table 11. Heterotrait-monotrait ratio of correlations ................................................... 49
Table 12. VIF - Inner models ....................................................................................... 50
Table 13. Hypotheses testing ....................................................................................... 51
Table 14. R-Square values ........................................................................................... 54
Table 15. f-square value ............................................................................................... 55
 1

ABSTRACT

Problem: Many universities in Ho Chi Minh City, Vietnam, are applying online
teaching methods along with online tests and assignments. After some tests, the authors
found out that the results are controversial, and they are unfair as dishonest behaviors
are under low control. Besides that, the traditional tests also witnessed numerous
cheating cases. Those actions occur more frequently along with the ease of access to AI
platforms. Reasons for choosing the topic: To have a better understanding of dishonest
behavior and the reasons for this, we decided to do this research “FACTORS
AFFECTING THE DISHONEST BEHAVIOR OF UNIVERSITY STUDENTS IN HO
CHI MINH CITY”. Hence, we can comprehend the factors leading to dishonest
behavior and recommend solutions for universities to manage this issue to bring out
better performances for the following tests. Methods: Based on the theoretical bases
given by the authors, we surveyed a sample of 313 university students from diverse
schools in Ho Chi Minh City. Results: The topic is based on the theoretical bases given
by the authors with key influencing factors such as Attitude towards cheating,
Subjective Norms, Horizontal-Vertical Individualism- Collectivism, and Intention,
which are strongly correlated with students’ cheating behavior. The remaining factors
are not regarded to have a significant impact on students' engagement. Conclusion:
Relied on that result, some recommendations and implications are proposed to improve
the quality of online tests or assignments and, hence, to enhance the performances of
university students.
Keywords: dishonest behavior, university students, Moral Obligation, HVIC, TPB.
 2

CHAPTER 1 - INTRODUCTION

1.1 Reasons for choosing the topic

1.1.1 Overview of the problem

Hendy, & Montargot, 2019, defined “academic dishonesty” as “any deviant

behavior taking place during an academic exercise”. Academic misconduct includes a
variety of dishonest activities when students cheat, plagiarize, or collude in their studies.
First of all, cheating is the unauthorized collaboration or the use of external resources
during exams, assignments, tests, etc., to achieve an unfair competitive advantage.
Moreover, plagiarism is another familiar type of academic misconduct. When students
carry out this, they submit work or projects that are not their own, passing it off as
original. Furthermore, ghost authorship happens when a student lets another person
write a paper on their behalf while faking or copying data. No matter what form of
cheating, plagiarism, ghost authorship, faking or copying the data, and so on, academic
dishonesty behavior has become all the rage across the world. Academic dishonesty has
been one of the most alarming concerns that universities must take into consideration.
The high popularity of academic dishonesty was confirmed in many famous studies
(McCabe & Trevino, 1993; McCabe & Trevino, 1996).

Not only does it seem to be unfair to those honest students doing hard work, but
it is also controversial as to why students need to go to school. In fact, should a student
misconduct, the academic environment will definitely become unfair. In other words,
academic dishonesty undermines the integrity of the academic institution and destroys
the value of degrees of students taking part in such misconduct activities. As a result, it
can be inferred that academic dishonesty will somehow devalue the entire education
system as a whole, because of the unfair advantage from those who obtain qualifications
through dishonest actions over those working against the clock and earning their
qualifications honestly. In the long run, this may lead to serious implications for the
students’ future careers and for society.

Despite the efforts to reduce academic misconduct, the problem still persists in
Vietnam. According to Uc et al., 2021, in Vietnam, notwithstanding generally receiving
lots of disapproval, academic misconduct is still a common practice in many institutions,
 3
in particular, at the university level. This may result from excessive workload, a shortage
of options in offered classes, and a heavy emphasis on examinations. Or, Vietnamese
students may feel under pressure due to high expectations from family, friends, society,
and even more, the educational system itself. Particularly, Ho Chi Minh City is regarded
as a bustling metropolis with a highly competitive educational place in Vietnam.
Students here, hence, somehow feel under such intense pressure to excel academically,
because they are psychologically manipulated that getting high grades means an indirect
ticket to have a better job with a higher salary, and the only way to get away from
poverty, that they choose to misconduct. As a consequence, Ho Chi Minh City is a
suitable case study for carrying out investigations on the factors contributing to
academic dishonesty behavior among university students.

Previous studies illustrate various factors associated with a student’s academic

dishonesty behavior. These factors are divided into two main components including
intentions and justifications which are affected by attitudes, subjective norms, and
perceived behavioral control (Thomas H. Stone et al. 2009).

It is highlighted that the shortage of awareness among students about the adverse
consequences and ethical considerations relates to academic dishonesty. A study by
Meital Amzalag et al. (2021) shed light on the fact that lecturers have negative attitudes
towards forms of academic misconduct. In fact, they showed their concerns about lower
academic performance, held a belief that conducive conditions lead to cheating, and
overestimated the potential of cheating. When it comes to students side, the researchers
found that students blame studying difficulties, or external factors like pressures from
peers or time constraints as the main reasons for engagement in academic dishonesty
and the tendency to “copy-paste" more while stating that cheating brings more benefits
than adverse consequences, which results in normalization and standardization of such
dishonest behavior, as mentioned by a study by Delvin & Gray (2007) on the growing
acceptance of plagiarism among students. Overall, those findings suggest that attitudes
towards cheating among both lecturers and students have a contribution to academic
dishonesty.

McCabe and Trevino (1997) argued that students often re-engage in unethical
decisions in the future. For instance, if a student cheats on an unimportant minor
assignment once, there is a great likelihood that they might cheat on a more important
 4
exam later on. This pattern shows that small acts of dishonesty can escalate to more
consequential acts of dishonesty performance later on (Garrett et al., 2016). The
negative impacts of this issue cannot be underestimated. Academic dishonesty can
diminish students’ personal morality along with academic success, and also build up a
repeated practice of such action in the educational system and workplace afterward.
Enrolling in such unethical behavior, additionally, can have adverse lasting effects on
students. Employers typically value those graduates who have a good academic
performance with ethical decision-making skills, hence, a history of dishonest behavior
can hinder future job opportunities.

Dang Hung Vu and Nguyen Thanh Long (2020) conducted a study on an

evaluation of student’s academic integrity by their perception of the academic
environment and dishonesty behaviors. They confirmed that in spite of the
implementation of strict school policies on academic integrity, a considerable number
of students still decide to continue committing cheating. The frequency of engaging in
academic dishonesty behavior is noted to be inversely correlated with students' low
awareness of such actions. It can be indicated that lacking in understanding the
seriousness and negative impacts of academic dishonesty, can proportionally contribute
to its commonness among university students. In response to this persistent issue, most
universities have announced stricter regulations in an effort to take control over cheating
behavior among students or even have invested much money in anti-plagiarism software
such as AI detectors, Turnitin,..., in order to catch and deter academic misconduct.
Although these efforts are made significantly, many students have successfully
committed fraud using increasingly sophisticated and modern means of cheating.

Last but not least, it is important to find the underlying causes of student
dishonesty. Universities should take the responsibility of making more efforts to create
a more supportive learning environment in which students feel less anxious. E.g.,
flexible learning choices, accessible studying support services, and less excessive
workloads. Moreover, open interaction between lecturers and students can help identify
difficult areas and then provide suitable support to students who are having academic
struggles.

In conclusion, there is no doubt that academic dishonesty among university

students in Ho Chi Minh City presents an alarming concern among lecturers,
 5
universities, and societies. Therefore, it is crucial to understand the context of this area
and the relationships of factors contributing to student dishonest behavior. The purpose
of this research is to contribute to a more comprehensive understanding of the problem
and help us to jump into effective solutions to support a culture of academic integrity
within Ho Chi Minh City's universities.

1.1.2 Necessity of the research

Some research studies (Ballantine et al, 2014; McCall, 1988; and Sims and
Sims, 1991) indicate that those who have unethical behavior at school also behave
unethically at the workplace. Researchers suggest that an individual’s academic
dishonesty behavior could tarnish the reputation of an entire community (Harding et al.,
2004). This is because of the socialization process in which one group member could
serve as a benchmark for another member’s action. Hence, acceptable behavior by
several members of the group is referred to by others as stereotypical behavior which
stands for the whole group.

There is a likelihood that the ones cheating at school will cheat in the workplace.
There are three main reasons for this. The first reason to mention is habitual dishonesty.
Cheating may become a behavioral pattern. Should someone escape from cheating at
school, they might be more willing to try it at work. Secondly, cheating weakens a
person's ethical code. Should someone justify cheating in one situation, they might be
more likely to justify it in others, eroding their overall sense of fairness and honesty,
which is called a moral compass. Last but not least, a permissive workplace environment
where cheating is ignored or even encouraged can make someone who already has a
propensity to cheat feel more comfortable doing so.

The implications of students’ willing engagement in academic dishonesty are far-

reaching for stakeholders in and beyond higher education. For example, research has
also shown that students who cheated in college were more likely to shoplift (Beck &
Ajzen, 1991), cheat on income taxes (Fass, 1990), abuse harmful substances
(Blankenship & Whitley, 2000), cheat in graduate and professional schooling (Baldwin,
Daugherty, Rowley, & Schwartz, 1996), and engage in unethical workplace behavior
(Harding, Carpenter, Finelli, & Passow, 2004; Hilbert, 1985; Nonis & Swift, 2001;
Ogilby, 1995; Sims, 1993; Todd-Mancillas, 1987).
 6
All of the aforementioned ideas explain why we embarked on the study
"FACTORS AFFECTING THE DISHONEST BEHAVIOR OF UNIVERSITY
STUDENTS IN HO CHI MINH CITY". This study will provide a closer look at
academic dishonesty and misconduct among students and the factors influencing the
intention to commit such actions. Depending on that, the school can come up with the
most optimal measures to provide a healthier learning and equal testing environment for
all students. Hence, a study on this problematic issue is urgently needed.

1.2 Research gap

Currently, there has been a lot of research that has focused on understanding the
factors that influence cheating behavior in the school environment, with classic theories
such as Ajzen's Theory of Planned Behavior (1991) serving as the foundation.
According to this theory, Attitude, Subjective Norms, and Perceived Behavioral Control
are the main factors that affect Intention and thereby affect Behavior. Previous studies
have demonstrated that most factors such as Attitude, Subjective Norms, and Perceived
Behavioral Control have a significant impact on the Intention and Justification of
cheating behavior (Hsiao & Yang, 2011; Stone et al., 2009; Yuslizaa et al., 2020). In the
majority of studies on academic dishonesty, students are required to disclose behaviors
they have previously engaged in. This implies that understanding how students justify
their cheating could strengthen the explanatory capacity of the Theory of Planned
Behavior (TPBC) model so Justification factors were also considered in the study.
Furthermore, Intention and Justification are also considered important mediating factors
leading to Cheating Behavior (Stone et al., 2009).

Because cheating behavior is increasing and becoming more intricate, many

scholars believe that this field needs to be researched more deeply to explore and
identify problems for timely resolution (Minh Đức Hồ, 2019). Most previous studies
have not considered moderating factors that may influence cheating decisions or
changes in justification. In particular, the role of factors such as Moral Obligation and
Horizontal-vertical Individualism-collectivism in moderating relationships has not been
fully studied. Recent studies have also begun to explore the impact of the two factors
Moral Obligation and Horizontal-vertical Individualism-collectivism on other factors
(Yinxia Zhang, 2023; Mayhew et al., 2009); However, the specific impact of these
factors on regulation remains unclear. Moral Obligation is thought to influence whether
 7
students cheat because it reflects personal feelings of moral responsibility in avoiding
misconduct (Lin & Wen, 2007; Tsui & Ngo, 2016). Meanwhile, Horizontal-vertical
Individualism-collectivism can shed light on how cultural values influence how
individuals justify cheating and translate intentions into actions (Hofstede, 2001).

Accordingly, this study aims to fill the current gap by investigating whether
Moral Obligation and Horizontal-vertical Individualism-collectivism factors moderate
the relationship between Intention, Justification, and Cheating Behavior. Results from
this study are expected to make an important contribution to a deeper understanding of
the regulatory mechanisms of cultural and ethical factors in cheating behavior, thereby
providing practical solutions for organizations and educational systems in managing and
minimizing cheating behavior through establishing ethical education programs and
building appropriate organizational culture.

1.3 Research questions

Through the problem analysis, the research team carries out a study answering
the following questions:

• What factors affect the dishonest behavior of university students in Ho Chi Minh City?

• Whether the intention to cheat actually lead to cheating behavior or not?

• What measures can educational institutions employ to preclude dishonesty and enhance
fairness in the academic environment?

1.4 Research objectives

The fundamental objective of the experiment is to expand the previous research

of (Matthew J. Mayhew et al, 2009). Simultaneously, the researchers look forward to
conducting these specific objectives:

1. Determine factors and evaluate the effect levels of those factors in the dishonest
behavior of university students in Ho Chi Minh City.

2. Propose some viable recommendations and implications for stakeholders in the field of
educational management to mitigate the negative effects and to curb the problem
properly.
 8
1.5 Research subjects & scope

• Research subject: Factors affecting the dishonest behavior of university students in Ho

Chi Minh City.

• Respondents: Students from various universities and colleges, who have (or have not)
engaged in academically dishonest behaviors.

• Research scope: The study is conducted by investigating university and college students
within Ho Chi Minh City exclusively.

• Data collection period: The study’s data collection was launched on June 9th, 2024, and
ended in the following month.

1.6 Research methodology

The study applied the Quantitative research method to gather data investigating
factors affecting the dishonest behavior of university students in Ho Chi Minh City. The
data collection process was carried out in the form of an online questionnaire using
Google Forms. Initially, the team adapted, filtered, and revised the questionnaire from
a multitude of relevant publications (mainly from studies of Matthew J. Mayhew et al,
2009; Thomas H. Stone, I.M. Jawahar, Jennifer L. Kisamore, 2009; Triandis, 1998), and
used a 5-point Likert scale ranging from strongly disagree to strongly agree for all the
measurement observed variables.

1.7 Research structure

The research is systematically designed as the following structure:

Chapter 1 - Introduction

Chapter 2 - Literature Review

Chapter 3 - Research Methodology

Chapter 4 - Data Analysis, Results, and Discussions

Chapter 5 - Conclusions and Implications

 9
1.8 Research contribution

1.8.1 Theoretical contributions

Along the way seeking to examine research questions, the study will dig deeper
in addition to prior research, which re-explores the role of further psychological factors.
It provides more insight into the current situation of dishonest academic practices among
students. The study highlights the complex factors influencing academic dishonesty in
Ho Chi Minh City universities and advocates for proactive measures to promote a
culture of integrity. It emphasizes the need for strict policies, an improved ethical
education curriculum, and a supportive environment to eliminate cheating behavior.
Future research should explore contextual factors and cultural influences, such as
workload design, assessment practices, and teaching styles, to gain a more nuanced
understanding of student dishonesty across different educational settings.

The study also suggests investigating the interplay between moral obligation and
horizontal-vertical individualism-collectivism to understand how cultural values
influence student decision-making regarding academic integrity. By acknowledging the
impact of intentions and justifications, universities can develop targeted interventions
and refine their understanding to promote a culture of academic integrity.

Besides, the thesis will be devoted to strengthening the theory of planned

behavior and Hofstede's cultural dimensions theory (in this study, only individualism-
collectivism is used to measure the effects on academic misconduct behaviors). With
the results obtained, it supports many previous publications and subsequently enhances
the validity of those theories in predicting dishonest practices among scholars.

1.8.2 Practical contributions

This research suggests that universities can effectively implement academic

codes of ethics by introducing them to faculty members. These codes should reflect the
university's perspective on integrity, organizational structure, trust, competitive
pressures, punishments, and clear rules. Prioritizing these codes at the university level
can foster a positive academic culture, influencing students' attitudes and perceived
behavioral control over cheating. Instructors can detect cheating by rearranging
classrooms and encouraging law-abiding behavior.
 10
Rewards based on class or quiz answers can also encourage diligent study.
Enforcing cheating regulations, lowering the cheating barrier, and highlighting the
honor code can change students' negative attitudes toward cheating. This is because
other students develop negative opinions about academic dishonesty when they witness
cheating being caught and punished. This can lead to more dedicated students resisting
the temptation to cheat.
 11
CHAPTER 2 - LITERATURE REVIEW AND HYPOTHESES
DEVELOPMENT

2.1 Relevant research

2.1.1 Using the theory of planned behavior and cheating justifications to predict
academic misconduct (Stone et al., 2009)

This article examines the usefulness of modifying Ajzen’s theory of planned

behavior to predict academic misconduct. The study extends the TPBC model in the
prediction of misconduct behavior. Attitudes, subjective norms, behavioral control,
intentions, and justifications were related to cheating behaviors. Academic misconduct
may be reduced by shaping attitudes toward cheating, changing perceptions of
subjective norms regarding the prevalence of cheating, and lowering students’
perceptions of their control of cheating by emphasizing the consequences of getting
caught.

This article is the research that we use as the main reference according to the trust
base - theory-driven. Theory-driven research is necessary to develop an understanding
of the rationale behind academic misconduct and to determine the most effective means
for curbing such behaviors (Stone et al., 2009). In Vietnam, there have been various
researches using TPB in suggesting research models (Hoang, 2023; Hoang, 2024).

However, there are still some limitations of this research. First, While the TPBC
(Ajzen, 1991) has been used to predict a wide variety of behaviors, its use has been
fairly limited in the area of academic misconduct (Stone et al., 2009).

Figure 1. Research model of Stone et al. (2009)

 12
2.1.2 Using structural equation modeling to validate the Theory of Planned
Behavior as a model for predicting student cheating (Matthew et al., 2009)

This article empirically validates the use of a modified form of the Theory of
Planned Behavior (TPBC) (Ajzen, 2002) as a model for predicting college cheating in
general and specifically for cheating by students in the developmental periods of
consolidation and transition. Besides that, they used the TPBC, Kohlbergian (1976)
notions of moral reasoning development, and the consolidation-transition model of
development advanced by Snyder and Feldman (1984) to ground an investigation of
certain psychological processes and behaviors and their subsequent effects on predicting
college cheating.

Figure 2. Research model of Matthew et al. (2009)

2.1.3 Academic cheating as planned behavior: the effects of perceived behavioral
control and individualism‐collectivism orientations (Yinxia, 2023)

This study investigates the moderating effect of perceived behavioral control

within the Theory of Planned Behavior (TPB) on academic dishonesty among university
students. While the TPB posits perceived behavioral control as a moderator, empirical
research in this context remains limited. The study extends this area by exploring the
influence of cultural dimensions, specifically focusing on the four facets of horizontal
and vertical individualism-collectivism. Through a multi-campus survey of 2,293
Chinese undergraduate students, the analysis reveals a significant negative interaction
 13
between perceived behavioral control and subjective norm. Interestingly, a positive but
non-significant interaction emerged between perceived behavioral control and attitude
toward cheating. Further analyses of these interactions suggest a pattern: the positive
effects of both attitude and subjective norm on academic dishonesty are amplified when
students exhibit low horizontal individualism and low horizontal collectivism,
respectively. These findings contribute to a more refined understanding of the TPB's
applicability in predicting academic cheating.

Figure 3. Research model of Yinxia (2023)

2.2 Key concepts

2.2.1 Term Definitions:

• Dishonest behavior

Academic dishonesty can be understood as intentional deception with personal

academic work as well as the work of others (Gaberson, 1997). Academically dishonest
behaviors are unpermitted ones applied to students by the university council to ensure
fairness and proper assessment of a student's abilities. Dishonest behaviors may include
cheating, lying, deception, fabrication, sabotage, impersonation, bribery, and plagiarism
(García et al., 2023).
 14

• Plagiarism

Presenting work or ideas from another source as your own, with or without
consent of the original author, by incorporating it into your work without full
acknowledgment. All published and unpublished material, whether in manuscript,
printed, or electronic form, is covered under this definition, as is the use of material
generated wholly or in part through the use of artificial intelligence (save the hen use of
AI for assessment has received prior authorization e.g. as a reasonable adjustment for a
student’s disability). Plagiarism can also include re-using your own work without
citation. Under the regulations for examinations, intentional or reckless plagiarism is a
disciplinary offense.”

• Peer pressure

Peer pressure refers to the influence exerted on someone usually a friend,

schoolmate, or anyone with a similar interest to do something or push them towards
something they may be resistant to do or may not otherwise choose to engage in. Peer
pressure often involves conformity, where the person changes their behavior, values, or
attitudes to align with those of their peers.

2.2.2 Variable Definitions

• Attitude (AT)

In psychology, an attitude refers to a set of emotions, beliefs, and behaviors

toward a particular object, person, thing, or event. Attitude can also be described as the
way we evaluate something or someone. For example, we tend to respond positively or
negatively about certain subjects.

• Subjective norms (SN)

Subjective norms are defined by Fishbein and Ajzen (1975, p. 302) as 'the
person's perception that most people who are important to him or her think he should or
should not perform the behavior in question' and the motivation to comply with these
expectations. This means that those with a strong sense of group identity (Johnston and
White, 2003) or fearful of social disapproval (Latimer and Martin Ginis, 2005) are more
likely to be influenced by subjective norms.
 15

• Perceived behavioral control (PBC)

It can be understood as one's "perceived control over the performance of a behavior"

(Ajzen, 2002, p. 668), which is the perception an individual has about the level of ease
or difficulty required to execute a task (Ajzen, 1991).

• Intention (IT)

Intention is closely akin to the concept of “motivation”, with high levels of

intention strength understood to represent strong motivation to perform a behavior.
Intentions play a prominent role in several theories of health behavior, including the
Theory of Reasoned Action (Fishbein & Ajzen, 1975), the Theory of Planned Behavior
(Ajzen & Madden, 1986), the Health Action Process Approach (Schwarzer, 2001), and
Temporal Self-regulation Theory (Hall & Fong, 2007).

• Justification (JT)

Nowhere has the critical impulse “overshot its target” as widely as in relation to
the concept and activity of justification (Latour, 2002). The truth of this proposition in
psychology is evidenced in the ambiguity of language and concepts dealing with the
truth of propositions generally: Reasons are not always reasonable, but often
“rationalizations”; moral justification might as easily be called “moralizing”; and what
is “just” can always be countered as just one’s opinion.

• Moral obligations (MO)

It is the requirement to pursue what we believe is right and act accordingly.

Unfortunately, while we would like to believe there are universal truths, there is no one
system of morality among humans. Moral obligations differ according to one's
morality.

• Horizontal-vertical Individualism-collectivism (HVIC)

Vertical collectivism includes perceiving the self as a part (or an aspect) of a

collective and accepting inequalities within the collective. Horizontal collectivism
includes perceiving the self as a part of the collective but seeing all members of the
collective as the same; thus, equality is stressed.
 16

• Cheating behavior (CB)

It may be considered a form of academic dishonesty: it is a way to present others’

academic work as one’s own interfering with the learning and the evaluation process, a
fraudulent means of achieving grades, being accompanied by the risk of detection and
punishment (Jensen et al., 2002).

2.3 Theoretical background

In this section, the authors use two main foundation theories to explain the hypotheses
development and the research model of this study.

• The theory of planned behavior: the relationship between behavioral intention

and actual behavior.

• Horizontal and Vertical Dimensions of Individualism and Collectivism:

2.3.1 The theory of planned behavior

The Theory of Planned Behavior (TPB) started as the Theory of Reasoned Action
in 1980 to predict an individual's intention to engage in a behavior at a specific time and
place. The goal of the idea was to account for every action that a person can control.
This model's central idea is behavioral intent, which is shaped by attitudes toward the
probability that a behavior will produce the desired result and subjective assessments of
the advantages and disadvantages of that result.

The theory of planned behavior (TPB) is a psychological theory that links beliefs
to behavior. The theory maintains that three core components including attitude (1),
subjective norms (2), and perceived behavioral control (3), together shape an
individual's behavioral intentions.

(1) attitudes toward the behavior, i.e. beliefs about behavior or its consequences;

(2) subjective norms, i.e. normative expectations of other people regarding the

behavior, and

(3) perceived behavioral control, i.e. the perceived difficulty or ease of

performing the behavior.
 17
Ajzen (2008) provides a bibliography of TPB articles on their website, which
includes 690 empirical pieces and 56 theory and review papers, demonstrating the broad
application of the theory. The TPB has been used to predict a wide range of behaviors,
according to a review of papers on Ajzen's website. Researchers have most frequently
utilized the TPB to predict actions about the promotion of health and safety as well as
environmental protection.

Although the TPB (Ajzen, 1991) has been applied to a broad range of behavior
prediction, its application to academic dishonesty has been rather limited. In particular,
it has been applied in two ex-post factors investigations of plagiarism and cheating
(Passow et al., 2006; Stone et al., 2007) as well as a priori study of lying, shoplifting,
and cheating (Beck and Ajzen, 1991). Additionally, the TPB was utilized in a review by
Whitley (1998) as the foundation for a model of academic misconduct, classifying
personality, demographic, and environmental characteristics. Beck and Ajzen's (1991)
study has been the most direct test of the TPB's predictive power for academic
dishonesty.

Despite not being included in the TPB model, Beck and Ajzen (1991)
incorporated moral duty as a predictor, claiming that it could provide incremental
validity to explain these kinds of actions. Although the results indicated that moral
obligation increased statistically significant explanatory power, Beck and Ajzen
concluded that this addition had little application. Whitley (1998), Passow et al. (2006),
and Harding et al. (2007) incorporated moral duty into their models despite this
consideration. According to Beck and Ajzen's (1991) findings, the greatest amount of
variation in lying and cheating may be explained by perceived behavioral control.
Cheating and lying are against rules and conventions and have not been as well
researched using the TPB as other behaviors such as safety, health, and conservationism,
which are generally beneficial and in line with societal norms.

By adding "perceived behavioral control," TPB can explain the relationship

between behavioral intention and actual behavior. Several studies found that TPB can
predict health-related behavioral intentions. TPB has improved the predictability of
intention in various health-related areas, including condom use, leisure, exercise, diet,
etc. In addition, TPB has helped to explain the individual's social behavior by including
social norms as an important contributing explanatory factor.
 18

2.3.2 Horizontal and Vertical Dimensions of Individualism and Collectivism

The study examines the cultural conditions that support this moderating role, with
a focus on the four horizontal-vertical individualism-collectivism orientations. Based on
multicampus survey data gathered from 2293 Chinese undergraduate students, the
moderation analyses indicate a significant and negative joint effect between perceived
behavioral control and the subjective norm construct, as well as a positive effect
between perceived behavioral control and the attitude toward cheating construct. By
comparing the four significant three-way interactions, further moderated moderation
analyses reveal a pattern: when a low subjective norm is combined with either low
horizontal individualism or low horizontal collectivism, the positive effects of attitude
toward cheating and subjective norm on academic cheating are stronger. The study
supports a multivariable intervention approach that closely integrates administrative
measures with students' attitudinal and normative beliefs, with a concern for the subtle
yet significant influences of cultural orientations. It also contributes to a more nuanced
understanding of the usefulness of the Theory of Planned Behavior in predicting
academic cheating.

As we previously reviewed, La Barbera and Ajzen (2021) called for empirical

investigations to test the extent to which the proposed moderating role of Perceived
behavioral control would be generalized to other behavioral domains and other cultural
contexts, given that the majority of the studies on the topic have been conducted in the
west. With an emphasis on the individualism-collectivism orientations, the study further
investigates the border circumstances under which the moderating effect of Perceived
behavioral control is (un)likely to promote knowledge of cheating via the TPB lens.

Individualism vs. collectivism has been discussed about academic dishonesty to

varying degrees. Notably, when looking for logical reasons for the variations in
cheating, researchers have frequently turned to the features of the opposing
individualism-collectivism backgrounds (Lin & Wen, 2007; McCabe et al., 2008).
However, only a few research have specifically examined the individualism-
collectivism orientations as variables in the study of cheating prediction (Martin, 2011;
Thomas, 2017; Zhang & Yin, 2020; Kasler et al., 2021). Current research appears to
 19
support the idea that there are cultural differences in cheating. However, no discernible
trend seems to suggest that individualistic or collectivistic orientations are more strongly
associated with academic cheating (or with which subtype). Our suggested connections
between the individualism-collectivism orientations and the TPB constructs in the
ethical domain of academic cheating are partially supported by the fact that both the
TPBC and individualism-collectivism theoretical formulations are based on one's
perceived significance of particular values. It seems sensible to hypothesize that the
cultural values that are ingrained in kids could serve as a compass, guiding them in
comprehending and adhering to standards of proper academic behavior.

2.4 Hypothesis development

Students with a negative attitude towards academic integrity (believing it's

unimportant, to be honest) will have a stronger intention to behave unethically. This
aligns with the Theory of Planned Behavior, which suggests attitudes lead to a stronger
intention to perform a specific behavior. A previous research (Thomas H. Stone, I.M.
Jawahar, Jennifer L. Kisamore, 2009) showed that attitudes were significantly related to
intentions. According to Matthew J. Mayhew, et al., 2013, students who are less
consistent in how they process information when reasoning about moral issues are more
likely to make decisions about cheating intention based on their attitude toward cheating
at any given time; for example, in one context, these students may experience cheating
as positive or thrilling but in another situation as negative or boring. As a result, the
authors hypothesize:

Hypothesis 1: Attitude has a positive impact on Intention.

When students have a positive attitude towards cheating owing to the lax rules
and policies against such action, they will have a tendency to strike up an intention to
cheat. Likewise, when the opportunity and ease to cheat is high, with no, or low risk of
being reprimanded if caught cheating, the students will intend to cheat. (Kam et al.’s,
2018).

Mustapha et al.’s (2016) study suggests that despite religious disposition or the
ease of cheating, attitudes towards cheating could be one of two leading indicators of
the intention to cheat. This study also indicates the implication of peer pressure, most
 20
especially in a collectivistic country like Malaysia. The need to belong to a group or
peer pushes students to cheat when they find some of their friends cheating.

As a consequence, the need exists to carry out more studies from a more holistic
perspective, including students from various religions. Based on the literature review,
the following hypothesis is posited:

Hypothesis 2: Attitude has a negative impact on Justification.

There is a positive linkage between externality and decision-making. Subjective

norms are a person's perception of the social expectations to adopt a particular behavior.
Subjective norm is influenced by a person's normative beliefs combined with the
person's motivation to comply. The previous research elucidated that entities are prone
to external factors without agreeing to them. In general, under the impact of major
tendencies, an entity is usually triggered by a sense of FOMO (Fear of Missing Out).
Hence, being within the vicinity of dishonest activity-engaged counterparts may give an
impulse to one’s intention gradually. Simultaneously, before doing something, people
will exhibit a greater tendency to justify such actions for themselves (Thomas H. Stone,
I.M. Jawahar, Jennifer L. Kisamore, 2009). The perception of misconduct within a
student's social group may affect how others react to such practices and ultimately
influence the likelihood of justification. For this reason, the research team proposes the
following hypotheses:

Hypothesis 3: Subjective norms have a positive impact on Intention.

Hypothesis 4: Subjective norms have a positive impact on Justification.

A person's view of how easy or difficult something is to accomplish is known as

perceived behavioral control, or ease of behavioral control. The ability to carry out the
behavior readily and the ability to anticipate potential barriers are both influenced by
prior experience (Ajzen, 1991). Stronger perceptions of behavioral control are found in
those who are highly confident in their skills and the opportunities that surround an
action (Ramdhani, 2011). Whether or not someone engages in academic dishonesty can
be determined by how seriously or lightly they believe the repercussions of their actions
will be taken. Stone, Jawahar, and Kisamore (2009) state that an individual who believes
the advantages of engaging in fraudulent activities outweigh the potential repercussions
or penalties will persist in their fraudulent activities.
 21
According to Whitley (1998), students are more likely to participate in dishonest
activity if they possess more proficient cheating techniques. Academic dishonesty is
frequently thought to depend on the chance to cheat in addition to abilities (Lin & Wen,
2007). The association between behavior and perceived behavioral control, both directly
and indirectly (via intentions), may be explained by an unpublished meta-analysis
(Cheung and Chan, 2000, as quoted in Ajzen, 2002). Cheung and Chan (2000)
discovered in their meta-analysis that, in addition to attitudes and beliefs, perceived self-
efficacy also significantly contributed to the variance in intentions and behaviors that
went beyond what was expected from intentions. Thus, the following is expected:

Hypothesis 5: Perceived behavioral control has a negative impact on Intention.

According to Thomas H. Stone, I.M. Jawahar, Jennifer L. Kisamore, 2009, when

a student has a high perceived behavioral control, they have a high confidence in
resisting cheating. Those students believing they can resist cheating are less likely to
form the intention to cheat. Otherwise, if they feel they lack the ability to resist cheating,
they might be more likely to form the intention to cheat. After engaging in such
academic dishonesty, they might resort to justification to manage the dissonance
between their action and their self-image as an honest person. They might conclude that
"The test was too hard so everyone cheats anyway," to lessen the cognitive discomfort.
Consequently, this study was undertaken to examine the negative effect of perceived
behavioral control on the justification of university students. Thus, we hypothesize:

Hypothesis 6: Perceived behavioral control is negatively related to Justification.

Since moral responsibility has been demonstrated to improve the theory of

planned behavior's predictive power (Ajzen, 1991), it is included as a variable in this
study because it turned out to be an essential factor in both models. Academic dishonesty
behavior and intentions to cheat have a strong connection (Harding, Mayhew, Finelli,
& Carpenter, 2007). Research has indicated that the intention to commit academic
dishonesty is a strong indicator of the actual behavior of academic dishonesty (Mayhew,
Hubbard, Finelli, & Harding, 2009). According to Keith-Spiegel and Whitley (2001),
academic dishonesty by students has ethical implications. Students who cheat on
examinations and assignments are more likely to receive higher grades than students
who do not cheat. As a result, when the test results and grade point averages of dishonest
 22
students are compared, honest students perform worse. The grades or scores will not be
real, and as a result, they cannot be trustworthy. Additionally, students learn that
cheating is acceptable when they witness others cheating and the institution does little
to discipline the perpetrators. Students could start to think that cheating is the best way
to succeed in college and that hard work does not lead to academic achievement. The
moral consequences that go unnoticed could impact students' decision to engage in
academic dishonesty. Therefore, we propose the hypothesis:

Hypothesis 7: Moral Obligation moderates the relationship between Intention and

Cheating Behavior

Moral Obligation is less subject to alter depending on the outcomes of particular

circumstances or situations in which the desired action occurs since it represents an
individual's context-free behavior evaluation. This variable was added to this
study since it has been shown to increase the theory of planned behavior's predictive
power (Ajzen, 1991). According to McCabe et al. (2002), justification has a favorable
effect on a student's cheating conduct. Study results by Chapman et al. (2004)
corroborate the false consensus effect. Thus, justifications will provide more variation
to the explanation of dishonest behavior than will the explanation of dishonest
intentions. According to Keith-Spiegel and Whitley (2001), academic dishonesty by
students has ethical implications. When individuals feel a strong moral obligation, their
likelihood of cheating decreases even if they have justifications for it. Conversely, when
moral obligation is weak, justifications are more likely to lead to cheating. In essence,
moral obligation acts as a buffer, moderating the relationship between the reasons
people give for their actions and their actual behavior. Therefore, we tentatively predict
the following:

Hypothesis 8: Moral Obligation moderates the relationship between Justification

and Cheating Behavior.

Individualism vs collectivism has been involved in academic dishonesty to

varying degrees. Remarkably, when seeking logical reasons for variations in cheating,
researchers have frequently turned to the features of the opposing individualism -
collectivism backgrounds (Lin & Wen, 2007; McCabe et al., 2008). By contrast, only
some studies (Martin, 2011; Thomas, 2017; Zhang & Yin, 2020; Kasler et al., 2021)
 23
have directly examined the individualism-collectivism orientations as investigated
variables in the prediction of cheating. Our research also appears to support the idea that
there are cultural variations in cheating. However, no discernible trend seems to suggest
that individualistic or collectivistic orientations are more strongly associated with
academic cheating. It makes sense to hypothesize that the cultural values that kids are
educated on could serve as a compass, guiding them in understanding and adhering to
standards of proper academic behavior. In particular, disparities in attitudes towards
academic dishonesty based on cultural backgrounds have been documented for AT. For
example, students from collectivistic cultures are reported to be more tolerant of
cheating behavior (e.g., cooperative cheating, when the student helps the other student)
than students from individualistic cultures (Lin & Wen, 2007; McCabe et al., 2008).
Then, we come up with the following hypotheses:

Hypothesis 9: Horizontal-vertical Individualism-collectivism moderates the

relationships between Intention and Cheating Behavior.

Hypothesis 10: Horizontal-vertical Individualism-collectivism moderates the

relationships between Justification and Cheating Behavior.

The positive linkage between intention and cheating behavior has been
mentioned in previous research. In the TPBC, intentions immediately precede behavior
and are regarded as a central factor in the model as they capture the motivation for
behavior (Ajzen, 1991; Beck and Ajzen, 1991). (McEachan et al., 2011) suppose that
behavioral goals or intentions, in turn, are strong predictors of behavior. Thus, the
authors hypothesize:

Hypothesis 11: Intention has a positive impact on Cheating Behavior.

Justification has a positive impact on a student's cheating behavior in a positive

way. McCabe et al. (2002) found the best predictor of cheating was students’
perceptions that their peers cheat. The false consensus effect has been supported by
studies conducted by Chapman et al. (2004). In their study, students significantly
underestimated the likelihood that other students would engage in various academic
misconduct behaviors when compared to the frequency with which they engaged in
similar acts. Chapman et al. (2004) found that the false consensus effect was more
pronounced for high than low-frequency cheaters. Thus, "others do it" becomes a
 24
rationalization, a cognition compatible with cheating. Justifications will add variance to
explaining cheating behavior over variance explained by cheating intentions. It is
hypothesized as:

Hypothesis 12: Justification has a positive impact on Cheating Behavior

The summary of the proposed research hypotheses is as follows:

Table 1. Summary of research hypotheses

No. Hypothesis Description

1 H1 Attitude is positively related to Intention

2 H2 Attitude is negatively related to Justification

3 H3 Subjective norms are positively related to the Intention

4 H4 Subjective norms are positively related to Justification

5 H5 Perceived Behavioral Control is positively related to Intention

6 H6 Perceived Behavioral Control is negatively related to Justification

7 Moral Obligation moderates the relationship between Intention

H7
and Cheating Behavior

8 Moral Obligation moderates the relationship between Justification

H8
and Cheating Behavior

9 Horizontal-Vertical Individualism-Collectivism moderates the

H9
relationship between Intention and Cheating Behavior

10 Horizontal-Vertical Individualism-Collectivism moderates the

H10
relationship between Justification and Cheating Behavior

11 H11 Intention is positively related to Cheating Behavior

12 H12 Justification is positively related to Cheating Behavior

Source: Authors’ summarization, 2024

 25
2.5 Proposed research model

Based on the above hypotheses and theoretical backgrounds, the proposed research
model is as follows:

Figure 4. Proposed research model

Source: Authors’ proposal, 2024
 26
CHAPTER 03 - RESEARCH METHODOLOGY

3.1. Procedure

In the phase, the process begins with developing the model and formulating
research hypotheses. This is followed by a series of steps. In order to identify topics for
investigation through the state of the art or academic journal papers posted in reputable
scientific journals where authors review existing literature extensively in order to point
out areas not adequately researched on leading to gaps that can be used for constructing
new research. The second step is conducting a comprehensive literature review of
theories and related studies aimed at establishing connections between variables. That
is, a theoretical framework was used to build research models and hypotheses. At this
point, the scales were completed and data collection was performed. The authors applied
some standard criteria when choosing these measurement scales as well as asking for
guidance from their professor, and also adjusted them according to Vietnam’s research
context. For this reason, it becomes crucial to confirm if draft scale definitions
correspond with those researched variables or not, hence a decision by authors to submit
it for appraisal before proceeding any further. The questionnaires were translated into
Vietnamese since the majority of respondents were Vietnamese so that they could be
easily understood by them besides giving accurate responses. This survey was
conducted by the authors on some group members so as not only to test its logic but also
to ensure good readability and ease of interpretation before being ultimately spread
through various social media platforms. Lastly, data analysis and conclusions will be
done once all data have been collected in full. In this study, the authors used quantitative
methods to evaluate measurement models and structural models, and the software used
to process and analyze data is SPSS 25.0 software and SmartPLS Version 4 software.
After the final results, the authors discuss, evaluate, and present the research findings,
thereby deriving implications for theory and practice, and recommendations for future
research.

3.2. Qualitative Research

In the process of behavioral research, qualitative research is often used. It is

conducted with the purpose of adjusting the observed variables of the scales to suit the
time, context, and research subjects. The group of authors implemented qualitative
 27
research methods through exchanges on social networking platforms with instructors
and discussions with groups of UEH students. Through those exchanges and
discussions, the group of authors filtered and added, and subtracted ideas and concepts
that were consistent with the original orientation and from there outlined the research
model. However, in the conjecture, there was a personal nature of the group of authors,
so this could cause some subjective errors in the research.

3.3. Quantitative Research

3.3.1. Research Subject

Students from universities in Ho Chi Minh City, from their first to sixth year or
in graduate education, have participated in end-of-term exams. The surveyed students
come from various faculties and different universities.

3.3.2. Sample Size

The proposed research model consists of 8 latent variables with 58 measured

variables. Consequently, the minimum size of the sample should equal 58*5 = 290
observable variables. The researchers decided to have their best results when using PLS-
SEM by collecting opinions from 327 participants. Therefore, after discarding those that
were not appropriate, there were a total of 313 observations.

3.3.3. Sample Selection and Data Collection Method

This study will implement two non-probability sampling methodologies:

Convenience sampling method: The authors created a questionnaire form based

on the Google Forms platform about “The factors affecting the dishonest behavior of
university students in Ho Chi Minh City” and posted it in groups on social media
platforms with a large number of students in Ho Chi Minh City, such as Facebook, Zalo,
Instagram, etc.

Snowball method: The authors distributed the questionnaire to their friends and
seniors at UEH and asked them to complete it and share it further with their peers at the
university.
 28
3.3.4. Data Analysis Method

Several factors led to the selection of the PLS-SEM methodology as the main
instrument for data analysis in this study. First, the PLS-SEM method was presented by
Hair et al. (2014) as a causal modeling strategy that maximizes the explained variance
of latent constructs. It considerably increases the variance explained in dependent
variables when compared to CB-SEM (Hair et al., 2017). Consequently, PLS-SEM is
thought to be more suitable, particularly in cases when researchers give the dependent
variable's predictive power priority (Henseler et al., 2009). PLS-SEM has certain
advantages, The CB-SEM method is parametric and therefore requires normal
distribution in your data and other restrictive assumptions. However, PLS-SEM is non-
parametric and much more flexible in satisfying the required assumptions. This
flexibility means that non-normal data distribution and skewness are not an issue for
PLS-SEM applications, nonmetric scales are more easily accommodated and therefore
more widely applicable (Hair et al., 2018). Furthermore, PLS-SEM permits
simultaneous estimation from the measurement model and the structural model,
preventing skewed or unsuitable portions of the estimate (Hair et al., 2018). In addition,
solutions containing as few as two lower (first) order constructs (LOCs) can be obtained
with ease using PLS-SEM. Furthermore, unlike with CB-SEM, it enables the
formulation of LOCs as formative for the HOC without compromising the model
definition (Sarstedt et al., 2019). Lastly, PLS-SEM rarely experiences problems with
convergence and can withstand extremely complicated models with hundreds of
observed variables (Hair et al., 2022).

As a result, we used PLS-SEM with SmartPLS Version 4 to analyze the data. The
analysis involves two key stages as described by Hair et al. (2019). First, we assessed
the measurement model to verify the reliability and validity of both reflective and
formative models. Next, we assessed the structural model after the measurement models
were validated, focusing on hypothesis testing and the relationships between variables.
SmartPLS 4, utilizing the PLS algorithm, was employed to analyze the accuracy of the
scales, including R^2 and f^2 values. Additionally, the significance of the path
coefficients was assessed using the bootstrapping method.
 29
3.4. Measurement scale

Eight key components of our study model - attitude, subjective norms, perceived
behavioral control, intention, justification, moral obligation, horizontal-vertical
individualism-collectivism, and cheating behavior - have been identified as essential to
our research framework. To guarantee the validity and reliability of our measurement
instruments, the variables are carefully chosen based on studies published in respectable,
high-ranking journals. They are then codified and standardized in accordance with the
study's setting.

A 5-item scale that was developed from the work of Stone et al. (2009) was used
to assess attitude. An example of a sample item is “I think reporting the cheating
behavior of other students is important”. Besides, a 5-item scale that was developed
from the work of Stone et al. (2009) was used to assess subjective norms. An example
of a sample item is “I suspected another student of cheating on exam in past year”. For
the examination of perceived behavioral control, we employed a 5-item scale adapted
from Stone et al. (2009). A sample item is, “I think cheating on a test or exam is easy”.
A 5-item scale that was modified from Stone et al. (2009) was used to assess intention.
An example of this is “I will turn in another’s work done as one’s own”. We investigated
justification using a 6-item Stone et al. (2009) scale. For example, “I am afraid of failing
the course”. Next, to assess moral obligation, we developed a 4-item scale from the work
of Mayhew et al. (2009). A sample item is, “Cheating on an in-class test or exam is
against my principles”. The evaluation of horizontal-vertical individualism-collectivism
incorporated a 16-item scale sourced from the research of Triandis, H. C., & Gelfand,
M. J. (1998). Take “When another person does better than I do, I get tense and aroused”
as a sample item. Lastly, a 10-item scale drawn from the research of Stone et al. (2009)
was used to assess cheating behavior. For example “I collaborated on assignment that
was supposed to be individual work”. The questionnaire uses a 5-point Likert scale
(With 1 = Strongly disagree, 2 = Disagree, 3 = Neutral, 4 = Agree, and 5 = Strongly
agree) to measure the variables most accurately.
 30
Table 2. Questionnaire

Sign Original items Adjusted items References

Attitude (AT)

AT1 Important to report cheating I don't think reporting the cheating Stone et al.
by other students. behavior of other students is important. (2009)

AT2 Always wrong to cheat. I don't think it is always wrong to cheat.

AT3 Report cheating by a I do not report cheating by a student whom

student whom I do not I know.
know.

AT4 Report cheating by a friend. I do not report cheating by a friend.

AT5 Report cheating is I don't think reporting cheating is

necessary for fairness to necessary to ensure fairness to honest
honest students. students.

Subjective norms (SN)

SN1 Suspected another student I suspected another student of cheating on Stone et al.
for cheating on exams in exams in past year. (2009)
past year.

SN2 Suspected another student I suspected another student of plagiarizing

of plagiarizing in past year. in past year.

SN3 Friends cheat and have not Friends cheat and have not been caught.
been caught.

SN4 Perceived frequency of I often encounter copying work between

inappropriate collaboration group members.
on assignments.
 31

SN5 Perceived frequency of I often encounter cheating in tests and

cheating during tests and exams.
examinations.

Perceived behavioral control (PBC)

PBC1 For me to cheat on a test or I think cheating on a test or exam is easy. Stone et al.
exam is easy. (2009)

PBC2 If I want to, I can cheat on a If I want to, I can cheat on a test or exam.
test or exam.

PBC3 I believe that I would have I believe that I would have a great deal of
a great deal of control over control over whether I get caught
whether I get caught attempting to cheat on an on-class or test
attempting to cheat on an exam.
on-class or test exam.

PBC4 I believe that I have the I believe that I have the skills needed to
skills needed to cheat on an cheat on an in-class test or exam in all
in-class test or exam in all circumstances.
circumstances.

PBC5 It is mostly up to me It is mostly up to me whether or not I

whether or not I successfully cheat on an in-class exam.
successfully cheat on an in-
class exam.

Intention (IT)

IT1 Turning in another’s work I will turn in another’s work done as one’s Stone et al.
done as one’s own. own. (2009)

IT2 Unapproved collaboration I will use unapproved collaboration on an

on an assignment. assignment.
 32

IT3 Getting test information I will get test information from a student
from a student who has who has taken it.
taken it.

IT4 Copying from someone on I will copy from someone on a test.

a test.

IT5 Using unapproved materials I will use unapproved materials on an

on an assignment. assignment.

Justification (JT)

JT1 Time pressure. I am under time pressure. Stone et al.

(2009)
JT2 Peer pressure. I am under peer pressure.

JT3 To increase my grade. I want to achieve high grades.

JT4 Other students do it. Other students do it.

JT5 Fear of failure. I am afraid of failing the course.

JT6 Laziness. I am lazy about studying.

Moral obligation (MO)

MO1 Cheating on an in-class test Cheating on an in-class test or exam is Mayhew et al.
or exam is against my against my principles. (2009)
principles.

MO2 I would feel guilty if I I would feel guilty if I cheated on an in-

cheated on an in-class test class test or exam.
or exam.

MO3 It would be morally wrong I think academic cheating is a violation of

for me to cheat on an in- ethics.
class test or exam.
 33

MO4 Being moral is my I have a responsibility to uphold good

responsibility. ethics.

Horizontal-vertical Individualism-collectivism (HVIC)

HVIC1 I'd rather depend on myself I'd rather depend on myself than others. Triandis, H. C.,
than others. & Gelfand, M.
J. (1998)
HVIC2 I rely on myself most of the I rely on myself most of the time; I rarely
time; I rarely rely on others. rely on others.

HVIC3 I often do "my own thing." I often do "my own thing."

HVIC4 My personal identity, My personal identity, independent of

independent of others, is others, is very important to me.
very important to me.

HVIC5 It is important that I do my It is important that I do my job better than

job better than others. others.

HVIC6 Winning is everything. I just need to win.

HVIC7 Competition is the law of Competition is the law of nature.

nature.

HVIC8 When another person does When another person does better than I do,
better than I do, I get tense I get tense and aroused.
and aroused.

HVIC9 If a coworker gets a prize, I If my friends achieve high results, I will

would feel proud. feel proud of them.

HVIC10 The well-being of my The happiness of my friends is important

coworkers is important to to me.
me.
 34

HVIC11 To me, pleasure is spending I enjoy spending time with others.

time with others.

HVIC12 I feel good when I I feel good when I cooperate with others.
cooperate with others.

HVIC13 Parents and children must Parents and children must stay together as
stay together as much as much as possible.
possible.

HVIC14 It is my duty to take care of It is my duty to take care of my family,

my family, even when I even when I have to sacrifice what I want.
have to sacrifice what I
want.

HVIC15 Family members should Family members should stick together, no

stick together, no matter matter what sacrifices are required.
what sacrifices are required.

HVIC16 It is important to me that I It is important to me that I respect the

respect the decisions made decisions made by my groups.
by my groups.

Cheating behavior (CB)

CB1 Copied a few sentences I copied a few sentences from a source but Stone et al.
from a source but not given not given credit. (2009)
credit.

CB2 Copied from another I copied from another student and turned
student and turned in as in as own.
own.

CB3 Helped someone cheat on a I helped someone cheat on a test.

test.
 35

CB4 Collaborated on assignment I collaborated on assignment that was

that was supposed to be supposed to be individual work.
individual work.

CB5 Turned in work done by I turned in work done by others.

others.

CB6 Copied from another I copied from another student on test.

student on test.

CB7 Used notes on test without I used notes on test without instructor
instructor permission. permission.

CB8 Received substantial help I received substantial help on assignment

on assignment without without permission.
permission.

CB9 Cheated on test in any way. I cheated on test in any way.

CB10 Used unfair methods to I used unfair methods to learn about a test.
learn about a test.

Source: Authors’ summarization, 2024

3.5. Data Analysis Process

3.5.1. Descriptive statistics analysis

We then began the process of pre-processing to verify and check the data set that
we collected from Google Forms. This involves making sure there are no missing data
in the data because their presence can lead to biased analysis results. Further, we also
determine whether any duplicate values exist in our dataset, and if they do, it may affect
our analyses. Redundant submission or data entry errors cause this, which impacts on
results of an analysis. That is why we have followed some procedures to remove such
records so that they become as concise as possible. Then, after assuring that the
information was clean and clear enough for use, it was ready for analysis. In order to
describe the demographics of the research sample, SPSS 25.0 software was employed
in this study. This versatile statistical tool allowed us to produce descriptive statistics
 36
and frequencies/percentages distributions for demographic variables too (Tavakol &
Dennick, 2011). The variables under consideration were broken into gender, university
name, major course undertaken by students, and current academic level.

3.5.2. Measurement Model

In this study, the authors evaluated the measurement model based on reliability
and validity. For reliability, we use factor loadings, Cronbach's Alpha reliability, and
Composite reliability coefficient (CR) to assess the reliability of the constructs being
studied. For validity, we use Average Variance Extracted (AVE) to evaluate convergent
validity, and The Fornell-Larcker criterion (Fornell et al., 1981) and the HTMT matrix
(Henseler et al., 2015) to check discriminant validity. Therefore validating the
measurement model in accordance with Hair et al.'s (2021) advice.

3.5.3. Assess the quality of observed variables

The purpose of assessing the quality of observed variables is to determine

whether any observed variables within a factor structure have a low capacity to explain
the parent latent variable. Next, we will remove weakly observed variables and retain
just those that well reflect the latent variable. The following method is provided by Hair
et al. (2017) in "A Primer on Partial Least Squares Structural Equation Modeling (PLS-
SEM)" to assess the quality of observed variables for measurement scales:

Examining the PLS-SEM algorithm to obtain the outer loading coefficient to

determine the significance of the result for an observed variable. According to Hair and
colleagues, an observed variable must have an outer loading coefficient of 0.7 or greater
in order to be deemed relevant. They recommend removing observed variables from the
model if their outer loadings are less than 0.4. The decision to keep or remove the
variable is based on the researcher's assessment when this coefficient is between 0.4 and
slightly below 0.7, taking into consideration additional indicators like composite
reliability (CR) and convergent validity (e.g., AVE coefficient) of that component.

3.5.4. Assessing Cronbach’s Alpha coefficient

According to Hair et al. (2022), the most common communication criterion for
assessing internal reliability is to use the Cronbach's system. This coefficient is defined
as follows:
 37

𝑀 𝛴𝑀 2
𝑖=1 𝑠𝑖
Cronbach’s 𝛼 = #𝑀−1$ ⋅ %1 − &
𝑠2𝑡

In this formula, 𝒔𝟐𝒊 is the variance of indicator variable i of a particular construct,

measured with M indicator variables (i = 1, …, M), and 𝑠#$ is the variance of the sum of
all M indicator variables of that construct.

Higher values of the Cronbach's coefficient, which runs from 0 to 1, indicate

higher construct dependability. In exploratory studies, values specifically between 0.60
and 0.70 are acceptable; in more advanced stages of research, values between 0.70 and
0.90 may be deemed satisfactory. Values that are higher than 0.90 (and even higher than
0.95), as they usually come from redundant items, almost repeating the same question
in the interview. Lastly, values less than 0.60 suggest poor internal consistency
reliability. According to DeVellis (2012), an Alpha coefficient exceeding 0.70 is
necessary for the reliability test of our study, with values nearer 1 being preferred.

3.5.5. Assessing Composite Reliability

Because of the limitations of Cronbach's coefficient α, it is necessary to apply

another measure of internal consistency reliability; that is, composite reliability (CR).
This factor is computed using the following formula, which accounts for the various
external loadings of the indicator variables:
+
& 𝛴*
'() '' (
CR = +
* *
&)'() '' ( * )'() ,-.(0' )

In this formula, 𝑙% is the standardized external factor loading of indicator variable

i of a particular construct, measured with M indicator variables (i = 1, …, M), 𝑒% is the
measurement error of indicator variable i, and 𝑣𝑎𝑟(𝑒% ) is the variance of the
measurement error defined as 1 - 𝑙%$ .

Comparable to Cronbach's α coefficient, the CR coefficient has a range of 0 to 1,

where larger values signify increased construct dependability. For exploratory research,
composite reliability between 0.6 and 0.7 is acceptable; between 0.7 and 0.9 is the ideal
range (Nunally & Bernstein, 1994). In cases where the Composite reliability value
exceeds 0.95, it is possible that there are many observed variables. The measurement
 38
criterion used in this study is Composite reliability (CR) greater than 0.7 (Hair et al.,
2013).

3.5.6. Assessing Convergent Validity

Convergent validity, according to Hair et al. (2022), is the degree to which one
measure positively correlates with another measure of the same construct. The average
variance extracted (AVE) is a frequently used metric to demonstrate convergent validity
at the construct level. The communality of a construct is equal to AVE. The formula for
this criterion is as follows:

2
𝛴 𝑀 𝑙𝑖
𝐴𝑉𝐸 = 3 𝑖=1 5
𝑀

Where 𝑙% is the standardized external factor loading of indicator variable i of a

particular outcome construct with M indicator variables. When the construct's average
AVE value is 0.50 or above, it means that it accounts for more than half of the variance
of its indicator variables. On the other hand, an AVE value of less than 0.50 signifies
that, on average, the variation explained by the construct is smaller than the variance
contained in the error terms of the indicator variables (Hock & Ringle, 2010). Therefore,
to weed out unimportant factors in our investigation, we shall apply the condition AVE
≥ 0.5.

3.5.7. Assessing Discriminant Validity

Discriminant validity, as defined by Hair et al. (2022), is the extent to which a

construct is truly different from other constructs according to empirical criteria. A
construct must be distinct and contain elements or information that are not represented
by other constructs in the model in order to be shown to have discriminant validity. The
Heterotrait-Monotrait (HTMT) and Fornell-Larcker criterion were utilized in this study
to evaluate discriminant validity.

For the Fornell-Larcker criterion, This technique compares the correlation

coefficients between a latent variable and other latent variables with the square root
AVE coefficient (SQRT(AVE)) of the observable variable scale used to measure a latent
variable. We infer that the scale ensures discrimination if the SQRT(AVE) coefficient
is greater than the remaining correlation coefficients (Fornell & Larcker, 1981).
 39
For the HTMT criterion, The better a scale's average correlation coefficient is,
the more it exceeds the average of its cross-correlation coefficients. The latent variable
shares more variation within a scale when the average correlation coefficient within that
scale is higher. The latent variable that was just discussed has less variation with other
hidden variables if the average of the cross-correlation coefficients is lower. By then,
the two latent variables' indicators will have attained discriminant validity. Henseler et
al. (2015) stated that if the HTMT index of a pair of factors is greater than 0.9, the
discriminant validity of the factor is violated. If the HTMT index is below 0.85, the
discriminant validity is well guaranteed. Therefore, we will take the threshold of 0.85
as the standard for the study.

3.6. Assessing Structural Model

The next stage is to assess the structural model's output after we have established
the validity and reliability of the structural measures. The goal of assessing the structural
model's output is to assist in ascertaining the model's capacity to explain one or more
relevant latent constructs or variables that are connected to the study goals. Typically,
this evaluation's contents consist of testing for multicollinearity, evaluating the path
coefficients to determine the significance and relevance of the links in the structural
model, and evaluating the structural model's explanatory capacity.

3.6.1. Assessing Multicollinearity

Multicollinearity, or the phenomenon where the independent variables in the

model depend on each other will occur if the independent variables are closely related
to one another, have a linear relationship, or are strongly correlated with one another.
Regression coefficients, p-values, and impact significance are all distorted when a
model exhibits collinearity or multicollinearity, which can lead to inaccurate inferences
about the nature of the relationship in the model. Hair et al. (2019), When to Use and
How to Report the Results of PLS-SEM. European Business Review provides the
following cutoff points for variance inflation factor (VIF) values for evaluating the
collinearity phenomenon:

• VIF ≥ 5: there is a very high possibility that collinearity exists and the model is
severely affected.

• 3 ≤ VIF ≤ 5: the model may be collinear.

 40
• VIF ≤ 3: the model does not have collinearity.

Therefore, in this study, we will rely on the above threshold values to evaluate
multicollinearity. If VIF is greater than 3, the authors will propose appropriate solutions
to reduce VIF to an acceptable threshold.

3.6.2. Relationship in structural model

Because PLS-SEM does not require that the data be normally distributed. The
parameterized tests used in the regression analysis to determine if the outer weight,
outer loading, and path coefficients are statistically significant cannot be performed
because there is no normal distribution. Consequently, to verify the significance level,
PLS-SEM employs a statistically significant coefficient based on its standard error,
acquired by bootstrapping. According to Hair et al. (2018), a return enlarged sample of
about 5,000 samples is suggested. We can get the experimental t-value and p-value for
each of the path systems in the structural model using the bootstrap standard error. The
test is statistically significant at the 5% level with a T-value > 1.96.

3.6.3. Assessing Coefficient of determination (R2)

The degree to which the independent variables in the model can explain a
dependent variable is shown by the R2 value. There will be an equal number of R2
coefficients in the model as there are dependent variables. Hair et al. (2017) believe that
it is difficult to give an empirical rule of thumb to accept the value of R2, which depends
on the complexity of the model and the research field. R2 varies between 0 and 1, where
a value closer to 1 indicates a high degree of explanation for the dependent variable and
a value closer to 0 indicates a poor level of explanation. But in order to determine the
R2 criterion for this investigation, the authors choose to refer to Henseler et al. (2009)'s
research piece. According to them, PLS path models with R2 values of 0.75, 0.50, or
0.25 are considered substantial, moderate, and weak, respectively. We chose to utilize
R2 adjusted because SmartPLS Version 4 offers an additional R2 adjusted coefficient in
addition to R2, which more precisely captures the explanatory power of independent
variables.
 41
3.6.4. Assessing Effect Size (f2)

Effect size f2 is a coefficient that evaluates the effectiveness of the impact of each
independent variable on the dependent variable, to see if the impact is strong, medium,
weak, or has no impact. In general, it aims to consider the importance of an independent
variable on the dependent variable.
 42
CHAPTER 4 - DATA ANALYSIS, RESULTS, AND DISCUSSIONS

4.1 Sample characteristics

Data for this study was collected in 2024, during which the research team
conducted a literature review, developed a questionnaire, and then conducted a survey
and collected information from the research subjects. In total, 327 responses were
received from the participants. However, to ensure the quality and validity of the data,
a thorough screening process was conducted.

This process involved evaluating the responses against specific criteria that were
set out in advance. After the screening process, 313 responses that fully met the
eligibility requirements were retained. These responses came from several universities
in Ho Chi Minh City, an area with a dynamic and diverse academic environment,
providing a suitable representative sample for the study. These responses will be used
to conduct detailed data analysis in the following sections of the study. Detailed
characteristics of the survey sample will be presented below:

Table 3. Gender

Frequency
Gender Percent (%)
(N=313)
Male 153 48,9
Female 160 51,1
Source: authors’ calculation, 2024

Of the total 313 students surveyed, there were 153 male students, accounting for
48.9%, and 160 female students, accounting for 51.1%.

Table 4. University and College

Frequency Percent
University/College
(N=313) (%)
University of Economics Ho Chi Minh City 78 24.9
Vietnam National University, Ho Chi Minh City 95 30.4
Foreign Trade University Ho Chi Minh City 25 8.0
Other 115 36.7
Source: authors’ calculation, 2024

Of the 313 students surveyed, 78 were from the University of Economics Ho Chi
Minh City (UEH), accounting for 24.9%. Students from the Vietnam National
 43
University Ho Chi Minh City (VNU-HCM) had the highest proportion with 95 students,
equivalent to 30.4%. Foreign Trade University Ho Chi Minh City (FTU-HCM) had 25
students participating, accounting for 8.0% of the total. Notably, there were 115 students
from other universities and colleges, accounting for 36.7%.

Table 5. Major

Frequency Percent
Major
(N=313) (%)
Economics 76 24.3
Accounting - Finance 46 14.7
Information Technology 30 9.6
Mechanics and Technology 27 8.6
Linguistics 29 9.3
Other 105 33.5
Source: authors’ calculation, 2024
Table 6. Academic year

Frequency Percent
Academic year
(N=313) (%)
Freshman 47 15.0
Sophomore 99 31.6
Junior 86 27.5
Senior 65 20.8
Other 16 5.1
Source: authors’ calculation, 2024

Of the 313 students surveyed, sophomores accounted for the highest proportion
with 99 students, or 31.6%. Third-year students accounted for 27.5%, while fourth-year
students accounted for 20.8%, which may indicate increased academic pressure and
preparation for graduation. Freshmen accounted for 15.0%, with 47 students, typically
new to the university environment, who may not have encountered many academic
challenges. Additionally, 16 students were in the “Other” group, accounting for 5.1%,
which included students who were not in traditional academic years or had a special
learning path.
 44
4.2 Descriptive statistics

Table 7. Descriptive statistics

N = 313 Min Max Mean Standard Deviation

Attitude (AT)
AT1 1 5 1.422 2.393
AT2 1 5 1.359 2.412
AT3 1 5 1.347 2.78
AT4 1 5 1.333 2.901
AT5 1 5 1.4 2.403
Subjective norms (SN)
SN1 1 5 1.239 3.342
SN2 1 5 1.232 3.444
SN3 1 5 1.295 3.428
SN4 1 5 1.233 3.489
SN5 1 5 1.244 3.511
Perceived behavioral control (PBC)
PBC1 1 5 1.136 3.272
PBC2 1 5 1.155 3.297
PBC3 1 5 1.232 3.217
PBC4 1 5 1.254 3.083
PBC5 1 5 1.217 3.121
Intention (IT)
IT1 1 5 1.364 2.936
IT2 1 5 1.342 2.962
IT3 1 5 1.321 3.246
IT4 1 5 1.333 3.026
IT5 1 5 1.339 3.051
Justification (JT)
JT1 1 5 1.399 3.192
JT2 1 5 1.393 3.23
JT3 1 5 1.405 3.454
JT4 1 5 1.355 3.182
JT5 1 5 1.358 3.396
JT6 1 5 1.454 2.939
Moral obligation (MO)
MO1 1 5 1.074 3.604
 45
MO2 1 5 1.043 3.751
MO3 1 5 1.061 3.764
MO4 1 5 1.044 3.824
Horizontal-Vertical Individualism-Collectivism (HVIC)
HVIC1 1 5 1.354 3.473
HVIC2 1 5 1.264 3.514
HVIC3 1 5 1.318 3.409
HVIC4 1 5 1.286 3.473
HVIC5 1 5 1.327 3.371
HVIC6 1 5 1.342 3.243
HVIC7 1 5 1.308 3.537
HVIC8 1 5 1.335 3.256
HVIC9 1 5 1.296 3.297
HVIC10 1 5 1.276 3.358
HVIC11 1 5 1.274 3.495
HVIC12 1 5 1.234 3.502
HVIC13 1 5 1.323 3.479
HVIC14 1 5 1.319 3.518
HVIC15 1 5 1.306 3.466
HVIC16 1 5 1.267 3.514
Cheating behavior (CB)
CB1 1 5 1.272 3.153
CB2 1 5 1.321 2.981
CB3 1 5 1.257 3.102
CB4 1 5 1.148 3.431
CB5 1 5 1.345 2.866
CB6 1 5 1.278 3.121
CB7 1 5 1.316 3.198
CB8 1 5 1.242 3.262
CB9 1 5 1.348 2.898
CB10 1 5 1.32 3.147
Source: authors’ calculation, 2024
4.3 Reliability and discriminant validity
All items with a factor loading less than 0.7 are eliminated following the initial
phase of factor analysis (Hair et al., 2017). Regarding the reliability test, good
measurement reliability is ensured by composite reliability (C.R.) and Cronbach's Alpha
 46
is more significant than 0.7 (Hair et al., 2013; DeVellis, 2012). According to Chin
(1998), convergent validity is also satisfied when the average variance extracted (AVE)
of all constructs is more than 0.5. The reliability metrics are displayed in table 8.
Table 8. Outer loadings
AT CB HVIC IT JT MO PBC SN
AT1 0.886
AT2 0.875
AT5 0.886
CB1 0.855
CB10 0.877
CB4 0.750
CB5 0.803
CB8 0.843
CB9 0.858
HVIC1 0.854
HVIC2 0.831
HVIC3 0.876
HVIC4 0.838
HVIC7 0.761
IT1 0.875
IT2 0.887
IT3 0.845
IT4 0.855
IT5 0.884
JT2 0.857
JT3 0.899
JT4 0.877
JT5 0.877
MO1 0.830
MO4 0.919
PBC1 0.764
PBC2 0.840
PBC3 0.834
PBC4 0.842
PBC5 0.731
SN1 0.857
SN2 0.850
SN3 0.872
SN4 0.820
SN5 0.856
 47
Note: AT= Attitude, CB= Cheating behavior, HVIC= Horizontal- vertical
Individualism- collectivism, IT= Intention, JT= Justification, MO= Moral Obligation,
PBC= Perceived behavioral control, SN= Subjective norms. All constructs were greater
than 0.7.
Source: authors’ calculation, 2024

According to Table 8, the testing results show that the outer loadings of all
constructs were greater than 0.7. Therefore, all items in this study are corrected and
acceptable.

According to Heseler, Ringle, and Sarstedt (2015), the Heterotrait-Monotrait

ratio (HTMT) and the Fornell-Larcker criterion were applied. The diagonal portions of
each construct, or the square root of the AVE, were shown to be more significant than
other inter-construct correlations in the sections that came next, supporting the
constructs' excellent discriminant validity. When all Heterotrait-Monotrait ratios of the
correlations between the constructs were less than the threshold of 0.85, the results of
HTMT also demonstrated that discriminant validity was not a problem for this study
(Henseler et al., 2015). It thereby strengthens the tested constructs' discriminant
validity.

To assess the reliability of the measures reflected in the study, indices such as
Cronbach's Alpha, Composite Reliability (CR), and Average Variance Extracted (AVE)
were used. According to researchers such as DeVellis and Thorpe (2021) as well as
Bagozzi and Yi (1988), Cronbach's Alpha value greater than 0.7 is considered to have
good reliability, indicating that the items in the scale are closely correlated with each
other and measure the same concept. In this study, all the structural indicators had
Cronbach's Alpha greater than 0.7, which indicates high internal consistency and
ensures that the measurements are reliable. CR, similar to Cronbach's Alpha, but takes
into account the loadings of each indicator, provides a more accurate view of reliability.
A CR value greater than 0.7, as noted in this study, is an indication of a highly reliable
measurement model.

The average variance extracted (AVE) is used to assess the extent to which the
observed variables explain the variance of the latent variables. An AVE greater than 0.5
indicates that most of the variance of the latent variable is explained by the observed
 48
variables, demonstrating the convergent validity of the scale. The AVE values in Table
9 are all greater than 0.5, indicating that the AVE also meets the requirement.

Table 9. Construct Reliability and Validity

Composite Composite
Cronbach's Average variance
reliability reliability
alpha extracted (AVE)
(rho_a) (rho_c)
AT 0.858 0.859 0.913 0.779
CB 0.911 0.913 0.931 0.692
HVIC 0.890 0.914 0.919 0.694
IT 0.919 0.921 0.939 0.756
JT 0.901 0.904 0.931 0.770
MO 0.703 0.761 0.867 0.766
PBC 0.862 0.871 0.901 0.645
SN 0.905 0.906 0.929 0.724
Note: AT= Attitude, CB= Cheating behavior, HVIC= Horizontal- vertical
Individualism- collectivism, IT= Intention, JT= Justification, MO= Moral Obligation,
PBC= Perceived behavioral control, SN= Subjective norms.
Source: authors’ calculation, 2024

The results showed that the discriminant validity was satisfactory, as shown by
the square root of each AVE being bigger than its highest bivariate correlations.

Table 10. Correlations between Research Constructs (Fornell - Larcker Criterion)

AT CB HVIC IT JT MO PBC SN
AT 0.882
CB 0.152 0.832
HVIC -0.644 -0.190 0.833
IT 0.148 0.814 -0.206 0.869
JT -0.471 0.092 0.582 0.022 0.878
MO -0.477 -0.069 0.514 -0.047 0.293 0.875
PBC 0.048 0.641 -0.055 0.690 0.040 0.047 0.803
SN -0.613 0.077 0.585 0.118 0.536 0.439 0.156 0.851
Note: AT= Attitude, CB= Cheating behavior, HVIC= Horizontal- vertical
Individualism- collectivism, IT= Intention, JT= Justification, MO= Moral Obligation,
PBC= Perceived behavioral control, SN= Subjective norms. Diagonal elements (in
bold) are the square root of the average variance extracted.
Source: authors’ calculations, 2024.
 49
Heseler, Ringle, and Sarstedt (2015) emphasized that discriminability is ensured
when the correlations between the constructs in the model are not too high, specifically,
the Heterotrait-Monotrait ratio (HTMT) of the correlations needs to be less than the
threshold of 0.85. Table ---- presents the HTMT values between the constructs. The fact
that all HTMT ratios are less than the threshold of 0.85 indicates that discriminability is
ensured. This means that the theoretical constructs used in the study do not overlap and
each construct measures a separate concept. As a result, the relationships tested in the
model are reliable and not affected by the ambiguity between different constructs.

Table 11. Heterotrait-monotrait ratio of correlations

HVIC HVIC MO x MO x
AT CB HVIC IT JT MO PBC SN
x JT x IT JT IT
AT
CB 0.171
HVIC 0.748 0.202
IT 0.165 0.888 0.221
JT 0.531 0.119 0.660 0.064
MO 0.607 0.096 0.650 0.071 0.352
PBC 0.097 0.720 0.120 0.767 0.093 0.108
SN 0.697 0.101 0.661 0.130 0.592 0.544 0.183
HVIC x JT 0.550 0.192 0.540 0.218 0.360 0.234 0.202 0.479
HVIC x IT 0.220 0.289 0.166 0.258 0.248 0.101 0.274 0.256 0.006
MO x JT 0.325 0.065 0.219 0.054 0.104 0.365 0.046 0.268 0.523 0.254
MO x IT 0.072 0.178 0.088 0.203 0.060 0.047 0.166 0.109 0.217 0.685 0.173
Note: AT= Attitude, CB= Cheating behavior, HVIC= Horizontal- vertical
Individualism- collectivism, IT= Intention, JT= Justification, MO= Moral Obligation,
PBC= Perceived behavioral control, SN= Subjective norms. All HTMT values are less
than 0.85, so discriminability is guaranteed (except for the IT value being less than 0,9
which can still be accepted).
Source: authors’ calculations, 2024.
 50
4.4 Common method bias

To test and rule out the possibility of CMB in the study, the authors followed the
guidelines of Podsakoff, MacKenzie, Lee, and Podsakoff (2003). One of the effective
methods to determine CMB is to check for multicollinearity through the Variance
Inflation Factor (VIF). VIF is an index used to measure the extent to which a predictor
variable is correlated with other predictor variables in the model. If the VIF values are
high, this may indicate severe multicollinearity and, therefore, the possibility of CMB.

In this study, the authors applied the method of assessing multicollinearity by

examining the VIF values generated for all constructs from PLS-SEM. According to
Ned (2015), a VIF value below the threshold of 3 is considered to have no
multicollinearity problem and therefore no signs of CMB. The results show that all VIF
values in the model are below the threshold of 3 (see Table 4.8). Ensuring the absence
of CMB is a key factor in maintaining the accuracy and reliability of the research model.

Table 12. VIF - Inner models

AT CB HVIC IT JT MO PBC SN
AT 1.660 1.660
CB
HVIC 2.650
IT 1.237
JT 1.613
MO 1.719
PBC 1.061 1.061
SN 1.697 1.697
HVIC x JT 2.067
HVIC x IT 2.884
MO x JT 1.704
MO x IT 2.462
Note: AT = Attitude, CB = Cheating behavior, HVIC = Horizontal-Vertical
Individualism-Collectivism, IT = Intention, JT = Justification, MO = Moral Obligation,
PBC = Perceived behavioral control, SN = Subjective norms. VIF values in the model
were well below the threshold of 3.
Source: authors’ calculations, 2024.
 51
4.5 Assessing Structural Model

4.5.1 Hypothesis testing

Following Hair, Ringle, and Sarstedt's (2011) approach, the authors tested the
model to check that the predicted path coefficients were stable. The PLS-SEM results
are displayed in Table 13.

Table 13. Hypotheses testing

Original Sample Standard
T statistics P Conclusion
Hypo Description sample mean deviation
(|O/STDEV|) values
(O) (M) (STDEV)
Attitude is positively related to
H1 0.200 0.199 0.065 3.063 0.002 Supported
Intention
Attitude is negatively related to
H2 -0.225 -0.224 0.068 3.320 0.001 Supported
Justification
Subjective norms are positively
H3 0.138 0.137 0.063 2.176 0.030 Supported
related to the Intention
Subjective norms are positively
H4 0.400 0.403 0.071 5.658 0.000 Supported
related to Justification
Perceived Behavioral Control is
H5 0.659 0.662 0.042 15.856 0.000 Supported
positively related to Intention
Perceived Behavioral Control is Not
H6 -0.011 -0.012 0.059 0.194 0.846
negatively related to Justification Supported
Moral Obligation moderates the
H7 relationship between Intention -0.110 -0.091 0.055 1.996 0.046 Supported
and Cheating Behavior
Moral Obligation moderates the
Not
H8 relationship between Justification -0.030 -0.027 0.037 0.821 0.412
Supported
and Cheating Behavior
HVIC moderates the relationship
H9 between Intention and Cheating 0.171 0.153 0.065 2.638 0.008 Supported
Behavior
HVIC moderates the relationship
Not
H10 between Justification and 0.043 0.039 0.041 1.041 0.298
Supported
Cheating Behavior
Intention is positively related to
H11 0.754 0.756 0.038 19.637 0.000 Supported
Cheating Behavior
Justification is positively related
H12 0.131 0.129 0.057 2.285 0.022 Supported
to Cheating Behavior
Source: authors’ calculations, 2024

The results of the structural equation modeling analysis showed that most of the
hypotheses in the study were supported, except for H6, H8, and H10. Specifically, the
 52
hypotheses regarding the relationship between subjective norms and justification (p =
0.000), perceived behavioral control and intention (p = 0.000), and intention and
cheating behavior (p = 0.000) were all confirmed. Students' attitudes also had a
significant impact on intention (p = 0.002) and justification (p = 0.001), indicating the
vital role of this factor in forming cheating intention and how students rationalize their
behavior. However, the hypothesis related to moral obligation, the moderating effects
of moral obligation on the relationship between justification and cheating behavior (p =
0.412), were not supported, indicating that moral factors are not strong enough to
prevent cheating behavior when students already have the justification. In addition,
perceived behavioral control did not affect justification (p = 0.846), and horizontal-
vertical individualism-collectivism did not affect the relationship between justification
and cheating behavior (p = 0.298). This suggests that cultural factors do not change the
way students justify their behavior.

This result emphasizes that in academic settings, factors such as attitudes,

behavioral control, social norms, and intentions play important roles in advocating
students' cheating behavior, while personal moral and cultural factors seem to have a
minor explanation.

4.5.2 Moderating variables

• Moral obligation

The results of the analysis show that moral obligation does have a moderating
effect on the relationship between intention and fraudulent behavior. Specifically, the
path coefficient for this relationship (MO x IT ® CB) is -0.110, with a p-value below
the threshold of 0.05, indicating that the effect is statistically significant. It indicates that
there exists an inverse effect of moral obligation on the relationship between intention
and fraudulent practices.

Nonetheless, Moral Obligation does not moderate the relationship between

Justification and Fraudulent Behavior. The path coefficient for this relationship (MO x
JT ® CB) is -0.030 with a p-value greater than 0.05, indicating that there is no
significant effect. The f-square value for this relationship was only 0.002, further
indicating that the effect of moral obligation on the relationship between justification
and cheating behavior was insignificant.
 53
These results suggest that, in the context of the study, moral obligation just plays
a significant role in preventing forming intention to conduct cheating behaviors.
Meanwhile, it does not have a strong enough moderating effect to influence students'
justification towards cheating behavior in this study.

• Horizontal-Vertical Individualism-Collectivism

The results of the analysis showed that the moderating effect of Horizontal-
Vertical Individualism-Collectivism on the relationships between Justification
Behavior, Intention, and Cheating Behavior was inconsistent. Specifically, the path
coefficient for the relationship between Justification Behavior and Cheating Behavior
(HVIC x JT ® CB) was 0.043 with a p-value greater than 0.05, indicating that
Horizontal-Vertical Individualism-Collectivism does not moderate this relationship.
The f-square value was only 0.004, indicating that the effect of HVIC on this
relationship was so small and almost insignificant. This suggests that whether students
tend to be individualistic or collectivistic in horizontal or vertical directions, it does not
change the way they justify and lead to cheating behavior. In contrast, when examining
the relationship between Intention and Cheating Behavior (HVIC x IT ® CB), the p-
value was 0.298, which was less than 0.05, indicating that Horizontal-Vertical
Individualism-Collectivism significantly moderated this relationship. However, the f-
square value was only 0.004, indicating that the moderation level of HVIC was
extremely low, and the actual impact on the relationship between intention and cheating
behavior was insignificant.

These results may suggest that in the research context, Horizontal-Vertical

Individualism-Collectivism does not play a significant role in regulating how students
justify cheating behavior, but it has a notable impact when students already have
cheating intentions. This may be related to the fact that the distinction between
individual and collectivism does not strongly influence students' final decisions when
they have formed cheating intentions. However, the level of regulation remains low,
suggesting that this cultural factor is not a major factor in preventing or promoting
fraudulent behavior.
 54

Figure 5. The PLS-SEM Results

4.5.3 Quality testing

According to Table 14, the result of R² values may be interpreted as follows. R²

can take any value between 0 to 1. Although the statistical measure provides some
valuable insights regarding the regression model, the user should not rely only on the
measure in assessing a statistical model. There is no universal rule on incorporating
statistical measures in assessing a model. The context of the experiment or forecast is
essential and, in different scenarios, the insights from the metric can vary.

Table 14. R-Square values

R-square R-square adjusted

CB 0.687 0.679
IT 0.500 0.495
JT 0.320 0.314

Note: CB = Cheating behavior, IT = Intention, JT = Justification.

Source: authors’ calculations, 2024.
 55
• The R² adjusted value of the dependent variable CB (Cheating Behavior) is 0.676.
Thus, the independent variables explain 67.6% of the CB (Cheating Behavior) variable
variation.

• The R² adjusted value of the dependent variable IT (Intention) is 0.495. Thus, the
independent variables explain 49.5% of the IT (Intention) variable variation.

• The R² adjusted value of the dependent variable JT (Justification) is 0.314. Thus,

the independent variables explain 31.4% of the JT (Justification) variable variation.

Table 15. f-square value

CB IT JT
AT 0.048 0.045
CB
HVIC 0.025
IT 1.470
JT 0.034
MO 0.001
PBC 0.817 0.000
SN 0.022 0.139
HVIC x JT 0.004
HVIC x IT 0.035
MO x JT 0.002
MO x IT 0.020
Note: AT = Attitude, CB = Cheating behavior, HVIC = Horizontal-Vertical
Individualism-Collectivism, IT = Intention, JT = Justification, MO = Moral Obligation,
PBC = Perceived behavioral control, SN = Subjective norms.
Source: authors’ calculations, 2024.

According to Table 15, the result of f-square (f²) values may be interpreted as
follows. Cohen’s f² is appropriate for calculating the effect size within a multiple
regression model in which the independent variable of interest and the dependent
variable are both continuous (Cohen, 2013).

• f square < 0.02 – very small (no impact):

• HVIC → CB - Horizontal-Vertical Individualism-Collectivism → Cheating

behavior (0.018).

• MO → CB - Moral Obligation → Cheating behavior (0).

 56
• PBC → JT - Perceived behavioral control → Justification (0).

• HIVC x JT → CB - Horizontal-Vertical Individualism-Collectivism x

Justification → Cheating behavior (0.004)

• MO x JT → CB - Moral Obligation x Justification → Cheating behavior (0.002)

• 0.02 ≤ f square < 0.15 - small impact:

• JT → CB - Justification → Cheating behavior (0.034).

• AT → IT - Attitude → Intention (0.048).

• SN → IT - Subjective Norms → Intention (0.022).

• AT → JT- Attitude → Justification (0.045).

• SN → JT - Subjective Norms → Justification (0.139).

• HIVC x IT → CB - Horizontal-Vertical Individualism-Collectivism x Intention

→ Cheating behavior (0.035)

• MO x IT → CB - Moral Obligation x Intention → Cheating behavior (0.020)

• f square ≥ 0.35 - strong impact:

• IT → CB - Intention → Cheating Behavior (1.470).

• PBC → IT - Perceived Behavioral Control → Intention (0.817)

4.6 Discussion

H1 - The analysis reveals a strong correlation between the intention to cheat and
an individual’s attitude toward cheating, aligning with previous research findings (Kam
et al., 2018; Mustapha et al., 2016; Chudzicka-Czupała et al., 2016; Uzun & Kilis, 2020).
According to Uzun and Kilis (2020), students' intentions to engage in plagiarism are
significantly influenced by their attitudes. If students perceive that cheating has minimal
or no real consequences, they might not feel compelled to report instances of cheating
among their peers, thus fostering a more permissive attitude towards dishonest
behaviors.

This perception that cheating lacks serious repercussions can reinforce the
likelihood of students engaging in such behaviors themselves. Consequently, addressing
and changing public perceptions about cheating is crucial. By emphasizing the negative
 57
consequences of cheating, educational institutions, and policymakers can alter attitudes
and reduce the inclination to cheat. Educating students about the long-term impact of
cheating on academic integrity and personal ethics can help shift their attitudes and
diminish the acceptance of dishonest practices.

H2 - The analysis indicates a significant positive relationship between Attitude and

Justification (O = -0.225, p = 0.001). This finding suggests that a more favorable attitude
towards cheating is strongly associated with a higher tendency to justify such behavior.

Specifically, the p-value of 0.001 confirms that this relationship is statistically

significant, meaning that the observed correlation is unlikely to be due to random
chance. The f-square value of 0.045 indicates a moderate effect size, suggesting that the
influence of Attitude on Justification is not only statistically significant but also
meaningful in practical terms. This positive relationship implies that when students hold
a more permissive or supportive attitude towards cheating, they are more likely to find
reasons or rationalizations to justify their own or others' dishonest behavior. For
example, students who view cheating as a minor or acceptable offense are more prone
to develop justifications for their actions, such as believing that "everyone does it" or
that the benefits of cheating outweigh the potential consequences.

H3 - The analysis reveals a significant positive relationship between Subjective

Norms and Intention to cheat (O = 0.138, p = 0.030). This finding indicates that students'
perceptions of what is socially acceptable or expected by their peers significantly
influence their own intentions to engage in academic dishonesty. The p-value of 0.030
confirms that this relationship is statistically significant, while the f-square value of
0.022 suggests a small to moderate effect size, indicating that Subjective Norms have a
meaningful impact on students' intentions to cheat.

This result is consistent with previous research by Beck and Ajzen (1991), Hendy
and Montargot (2019), Jalilian et al. (2016), and Mustapha et al. (2016), which also
found that subjective norms play a crucial role in shaping individuals' intentions and
behaviors. When students perceive that cheating is normalized or accepted within their
social circles, they are more likely to intend to cheat themselves. The positive
relationship implies that if students believe their peers view academic dishonesty as
 58
acceptable or if they perceive that there is little disapproval from their social
environment, they are more likely to form the intention to cheat.

H4 - The analysis shows a significant positive relationship between Subjective

Norms and Justification (O = 0.400, p = 0.000). This finding indicates that when students
perceive that their peers approve of or are indifferent to cheating, they are more likely
to justify their own dishonest behavior. The p-value of 0 confirms that this relationship
is statistically significant, suggesting that the observed effect is unlikely to be due to
chance. The f-square value of 0.022, while indicating a small to moderate effect size,
reveals that the influence of Subjective Norms on Justification is meaningful in practice.

The positive relationship implies that students are influenced by their social
environment when rationalizing academic dishonesty. Factors such as peer pressure, the
desire to help friends, and the perception that others are also engaging in academic
misconduct can contribute to a student's justification for cheating. Additionally, fears of
failure, extenuating circumstances, and the absence of stringent anti-cheating policies
further exacerbate this tendency.

H5 - The analysis reveals a strong positive relationship between Perceived

Behavioral Control and Intention to cheat (O = 0.659, p = 0.000). This indicates that
students who believe they have a high degree of control over their ability to cheat are
significantly more likely to intend to engage in academic dishonesty.

The p-value of 0.000 confirms that this relationship is highly statistically

significant, while the f-square value of 0.817 indicates an exceptionally large effect size,
demonstrating a strong impact of Perceived Behavioral Control on Intention. This
finding aligns with the research of Jalilian et al. (2016), Kam et al. (2018), and Lonsdale
(2017), all of which highlight the critical role of perceived control in shaping intentions
to cheat. For instance, Kam et al. (2018) found that in Hong Kong, students’ perceptions
of their control over cheating significantly influenced their intention to engage in
dishonest behavior.

The results suggest that students are more likely to form the intention to cheat if
they perceive that cheating is relatively easy to execute and that there are minimal
repercussions for being caught. When students believe they have the means and
opportunity to cheat with minor risks of severe consequences, their intention to engage
 59
in such behavior increases. This perception may arise from situations where cheating is
facilitated by lax supervision, inadequate testing procedures, or a lack of stringent
enforcement of academic integrity policies.

H6 - The analysis shows that Perceived Behavioral Control is negatively related

to Justification (O = -0.011, p = 0.846). This result indicates that there is no significant
relationship between students' perceived control over their ability to cheat and their
tendency to justify dishonest behavior. The p-value of 0.846 suggests that this
relationship is not statistically significant, and the f-square value of 0.000 indicates that
the effect size is negligible. This lack of significant correlation may be due to differences
in the conceptualization and measurement of justification in this study compared to
previous research, such as Stone et al. (2009).

In this study, the justification measure included factors such as parental support
and friends' support, which were not considered in Stone et al. (2009). This variation in
content domains could account for the observed difference in the relationship between
Perceived Behavioral Control and Justification. Specifically, the inclusion of social
support elements might influence how students justify their behavior differently than
the factors considered in earlier research.

H7 - The analysis reveals that Moral Obligation moderates the relationship

between Intention and Cheating Behavior (O = -0.110, p = 0.046). This finding suggests
that Moral Obligation plays a role in influencing how students' intentions to cheat
translate into actual cheating behavior. The p-value of 0.046 indicates that this
moderating effect is statistically significant, albeit with a small effect size, as indicated
by the f-square value of 0.002. This result is consistent with the conclusions of Yayra
Dzakadzie (2021), who found that Moral Obligation moderates the intention toward
academic dishonesty. Specifically, Dzakadzie's research demonstrated an inverse
relationship between Moral Obligation and the intention to cheat, meaning that as
students' sense of moral obligation increases, their intention to engage in academic
dishonesty decreases.

The presence of Moral Obligation as a moderating factor suggests that students

with a stronger sense of moral duty are less likely to act on their intentions to cheat. This
moderation effect underscores the importance of moral values in shaping ethical
 60
behavior. When students perceive a strong moral obligation, they are more likely to
resist the temptation to cheat, even if they initially intend to.

Considering these findings, educational stakeholders should consider integrating

and emphasizing moral and ethical values within academic institutions. By fostering a
culture that highlights the importance of integrity and personal responsibility,
institutions can help strengthen students' moral convictions and thereby reduce the
likelihood of cheating. Implementing programs and initiatives that reinforce ethical
behavior can contribute to a fairer and healthier educational environment.

H8 - The analysis indicates that Moral Obligation does not moderate the
relationship between Justification and Cheating Behavior (O = -0.030, p = 0.412). This
result suggests that Moral Obligation does not significantly influence how students
justify their cheating behavior.

The p-value of 0.412 reveals that the moderating effect of Moral Obligation is not
statistically significant, and the f-square value of 0.002 indicates a negligible effect size.
This finding implies that Moral Obligation does not significantly alter the relationship
between Justification and Cheating Behavior.

Like Hypothesis 7, this result suggests that Moral Obligation has an inverse
influence on justification for academic dishonesty. In this context, the influence of moral
values on justification may be underestimated, resulting in a lack of significant impact.
This could be due to a range of factors, including the subjective nature of the
respondents' moral development, cultural background, individual experiences, and the
extent of ethical education they have received.

The lack of a moderating effect suggests that students' perceptions of moral

obligation do not have a substantial effect on how they justify their academic
misconduct. This might be because individuals interpret and apply moral values
differently based on their unique personal and cultural contexts. For instance, students
from various backgrounds may have varying levels of moral development or ethical
training, which can lead to diverse attitudes toward justification and dishonesty.

H9 - The results reveal that Horizontal-Vertical Individualism moderates the

relationship between Intention and Cheating Behavior (O = 0.171; p = 0.008). Although
the moderation effect is relatively small, it indicates a proportional impact, meaning that
 61
Horizontal-Vertical Individualism enhances the intention toward dishonest behaviors.
This effect is best explained by Vertical Individualism, which emphasizes the essence
of rivalry.

Vertical Individualism captures the spirit of competition, and in this context,

students with prominent levels of competitiveness are more prone to engage in
misconduct to achieve academic success. Once an intention to cheat is formed, the drive
associated with Vertical Individualism provides additional impetus, making dishonest
behavior more likely. Students who are highly competitive may resort to unethical
practices as a means to attain higher academic accomplishments, especially when they
are already inclined to cheat. The f-square value of 0.021, although small, still indicates
a notable impact in amplifying the tendency towards cheating when Vertical
Individualism is activated. This highlights the role of a competitive environment and the
psychology of rivalry in promoting cheating behaviors among individuals who prioritize
personal success more than anything else.

H10 - The analysis reveals that Horizontal-Vertical Individualism does not

moderate the relationship between Justification and Cheating Behavior (O = 0.043, p =
0.298). This suggests that Horizontal-Vertical Individualism does not significantly
influence how students' justification for cheating impacts their actual cheating behavior.
The p-value of 0.298 indicates that the moderating effect is not statistically significant,
and the f-square value of 0.004 suggests an exceedingly small effect size. This result
implies that Horizontal-Vertical Individualism, while it does influence cheating
behavior overall, does not have a significant role in moderating the relationship between
justification and cheating.

One possible explanation for this finding is that horizontal-versed individualism is

based on internal factors such as personal ego and self-reliance, which may not directly
intersect with the external factors that typically drive justification for cheating.
Justification for cheating often arises from external influences like observing peers'
cheating behaviors or convincing oneself that cheating is acceptable under certain
circumstances. In contrast, Horizontal-Vertical Individualism is characterized by
internal motivations and self-reliance, which are less connected to external justifications
for dishonest behavior.
 62
The horizontal individualism factor, which emphasizes independence and personal
ego, might not align with the way students rationalize or justify their cheating behaviors.
Since justification is strongly influenced by external social factors and perceptions of
others' behaviors, it operates independently of the internal, ego-driven motivations
associated with Horizontal-Vertical Individualism.

H11 - The analysis demonstrates that Intention has a strong positive impact on
Cheating Behavior (O = 0.754, p = 0.000). This indicates a highly significant and
substantial effect of intention on the likelihood of engaging in cheating behavior. The
p-value of 0.000 confirms that this relationship is statistically significant, while the f-
square value of 1.491 indicates an exceptionally large effect size. This finding aligns
with previous research, including studies by Ajzen (1991) and Beck and Ajzen (1991),
as well as McEachan et al. (2011). These studies support the notion that intention is a
critical predictor of behavior.

According to the Theory of Planned Behavior, intentions capture the motivational

factors that drive individuals to act. They reflect the extent to which people are willing
to put in effort and attempt to perform a specific behavior. As McEachan et al. (2011)
note, intentions are crucial indicators of how hard individuals are prepared to try to
achieve their goals. The stronger the intention to engage in a behavior, the higher the
likelihood that the behavior will be executed.

In the context of academic dishonesty, students who have a strong intention to

cheat are more likely to follow through with cheating behaviors. This is because
intention serves as a precursor to actual actions, guiding and influencing the effort and
determination students will exert in carrying out dishonest activities.

H12 - The analysis reveals that Justification positively affects Cheating Behavior
(O = 0.131; p = 0.022), although with a relatively small impact (f-square = 0.032). This
result indicates that the process of justifying cheating behavior does have a statistically
significant influence on whether students engage in academic dishonesty, but the effect
size is modest.

According to McCabe et al. (2002), peers' behaviors can provide normative support
for cheating, making it seem like an acceptable strategy for achieving academic success.
When students observe their peers engaging in dishonest practices, they may come to
 63
view cheating as a legitimate or even necessary action to gain or maintain an academic
advantage. This peer influence creates a social context in which cheating is normalized,
thereby reducing the perceived severity of dishonest behaviors.

Further supporting this notion, McCabe (1992) suggests that neutralization

strategies play a significant role in the relationship between justification and cheating.
Students often use these strategies to rationalize their own dishonest actions by
observing that others are cheating without facing repercussions. This external
observation can lead students to believe that cheating is permissible or justified,
especially when institutional or faculty responses to cheating are perceived as
inadequate.

In essence, when students see others around them engaging in academic

dishonesty, they receive an implicit form of validation that can reduce their internal
moral constraints against cheating. This external justification diminishes the perceived
severity of cheating, making it easier for students to convince themselves that their own
dishonest behavior is acceptable.
 64
CHAPTER 5 – CONCLUSIONS AND IMLICATIONS

5.1 Discussion

This study examined several significant factors that could lead to Ho Chi Minh
City University students’ intention to cheat. It demonstrated that students' opinions
toward cheating have an enormous impact on whether they opt to cheat or not, similar
to other studies (Kam et al., 2018; Mustapha et al., 2016; Chudzicka-Czupała et al.,
2016; Uzun & Kilis, 2020). Students who think cheating is an obvious action or believe
they won't face any serious consequences are more prone to cheat. This means we should
change how people think about cheating by emphasizing the seriousness of academic
dishonesty (Uzun & Kilis, 2020). Additionally, students with positive attitudes toward
cheating are more likely to justify their dishonest behavior, according to our research
(H2) (Soloshonok & Shmeleva, 2019; Stone et al., 2009).2009). Additionally, we
discovered that peer pressure and social standards play a big role in students' decisions
to cheat. If students think their friends or society accept cheating, they are more likely
to justify it (H4) and plan to do it (H3) (Beck and Ajzen, 1991; Hendy and Montargot,
2019; Jalilian et al., 2016). We found that students who believe they can cheat without
getting caught are more likely to decide to cheat (H5) (Jalilian et al., 2016; Kam et al.,
2018). On the other hand, having strong moral values can decrease the chances of
students cheating, even if they intend to do so (H7) (Yayra Dzakadzie, 2021).
Interestingly, while having strong morals didn't affect the association between having
excuses and cheating (H8), a competitive and individualistic mindset did influence the
intention to cheat (H9). This shows that students who desire value competition and
individual achievement are more likely to cheat to achieve academic success (Hofstede,
2001). Our study confirmed that students’ intentions have a strong impact on their
engagement in cheating behavior (H11), as suggested in previous research (Ajzen, 1991;
Beck and Ajzen, 1991; McEachan et al., 2011 who established intention as a key
determinant within the Theory of Planned Behavior (TPBC)). This highlights the
importance of addressing the root causes that lead students to form the intention to cheat
in the first place. Moreover, justifications for cheating (H12) were found to have a
modest impact on cheating behavior, influenced by peer behavior and strategies that
normalize academic misconduct among students (McCabe et al., 2002). This
 65
underscores the need to foster a culture of academic integrity where ethical conduct is
the norm and students are discouraged from rationalizing dishonest behavior.

5.2 Research contributions

5.2.1 Theoretical contributions

To effectively address the situation of academic dishonesty in Ho Chi Minh City

requires comprehensive solutions, emphasizing on attitudes, subjective norms,
perceived behavioral control, and moral obligations among students. Universities need
to enforce strict policies, improve ethical education curricula, and foster a supportive
studying environment to eliminate cheating behavior as much as possible. In brief, our
study highlighted the intricate factors influencing academic dishonesty while advocating
proactive measures to promote a culture of integrity in higher education.

In conclusion, our study provides a foundation for tackling the complex issue of
academic dishonesty in Ho Chi Minh City universities. By acknowledging the
significant impact of intentions and justifications, as well as the importance of the TPBC
model, universities can develop targeted interventions. Furthermore, by encouraging
future research to explore contextual factors, cultural influences, and the moderating
effects of moral obligation and HVC, we can continuously refine our understanding and
develop even more effective strategies to promote a culture of academic integrity within
higher education. Ultimately, by consistently expanding our knowledge and efforts to
combat the multifaceted aspects of academic dishonesty, we can ensure just and ethical
teaching methods that uphold integrity in both academic communities and students.

5.2.2 Pratical implications

In terms of practical implications, this research provides valuable insights that

many parties, especially universities management, academic staff, and policymakers
can use.

Academic codes of ethics can be introduced by faculty members, and they can
successfully convey to students the importance of abiding by these codes. A faculty's
perspective regarding integrity, organizational structure, a trusting environment,
competitive pressures, the harshness of punishments, the presence of clear rules
addressing improper behavior, and faculty monitoring should all be explicitly reflected
 66
in the academic codes of ethics. This academic code of ethics should continuously be
prioritized at the university level as a shared value to foster a great academic culture
that can significantly affect students' attitudes, subjective norms, and perceived
behavioral control over their intention to cheat.

If instructors prioritize enforcing academic rules of ethics and rearranging

classrooms, they will be able to detect instances of cheating by students during exams.
Because there is a significant probability of being caught, these students are unlikely
to cheat, hence they will have little opportunity to do so. Encouraging students to be
law-abiding and conscientious will also ensure that they do not feel compelled to
cheat. Lecturers may give out rewards to students based on how well they answer
questions during class or on pop quizzes, which could encourage students to work
more and more diligently in their studies.

It is possible to alter students' thoughts and ideas about the negative attitude
toward cheating on campus by enforcing the regulations against cheating, lowering the
barrier to cheating, and highlighting the honor code. This is primarily due to the fact
that other students develop negative opinions about academic cheating when they
witness students who plan to cheat being caught and given harsh punishment. As a
result, industrious and dedicated students would develop negative opinions about
academic dishonesty and be more resistant to the temptation to cheat.

5.3 Limitations and Further Research

Constructive as this research may be, there remain some shortcomings during the
execution procedure. In the first place, the study subject is deemed to be sensitive to
students, thus prompting adversities in recording quality responses and data of the
survey. This limitation causes only a few groups of students to be willing to join the
questionnaire; consequently, the lack of diversity in university brand engagement is
inevitable. Besides, the research only studies within a small geographic location (Ho
Chi Minh City), which may affect the generality of the outcome. Last but not least, the
concept of dishonest behaviors is a compound of many activities comprising cheating,
plagiarism, copying, unofficial use of publications, falsification of data, fabrication of
citations, and more. This study mainly puts emphasis on cheating, plagiarism, unofficial
 67
use of publications, and academic impersonation as a whole while neglecting other
components.

Future research should delve deeper into additional contextual factors and
cultural influences that impact cheating behaviors. By delving deeper into contextual
factors and cultural influences, we can gain a more nuanced understanding of student
dishonesty across various educational settings. For instance, exploring the impact of
workload design, assessment practices, and teaching styles on student perceptions and
justifications for cheating can provide valuable insights. Additionally, investigating the
interplay between moral obligation and horizontal-vertical individualism-collectivism
(HVIC) can offer further insights into how cultural values influence student decision-
making regarding academic integrity.
 68
REFERENCES

(N.d.). Choosingtherapy.com. Retrieved August 24, 2024, from

https://www.choosingtherapy.com/peer-pressure/
Abu-Shanab, E., Al-Sharafi, M. A., & Al-Emran, M. (2023). The Influence of Network
Externality and Fear of Missing out on the Continuous Use of Social Networks: A
Cross-Country Comparison. International Journal of Human-Computer
Interaction, 40(15), 4058–4070. https://doi.org/10.1080/10447318.2023.2208990
Ajzen, I. (1991). The theory of planned behavior.Organizational Behavior and Human
Decision Processes, 50(2), 179-211. https://doi.org/10.1016/0749-5978(91)90020-t
Ajzen, I. (2002). Attitudes, personality, and behavior. Buckingham, England: Open
University Press.
Ajzen, I. (2002). Perceived behavioral control, self‐efficacy, locus of control, and the
theory of planned behavior 1. Journal of applied social psychology, 32(4), 665-683.
Ajzen, I. (2008). Consumer Attitudes and Behavior. Dalam Handbook of Consumer
Psychology, 525-548. Lawrence Erlbaum Associates.
Ajzen, I., & Madden, T. J. (1986). Prediction of goal-directed behavior: Attitudes,
intentions, and perceived behavioral control. Journal of Experimental Social
Psychology, 22, 453–474. https://doi.org/10.1016/0022-1031(86)90045-4
Al Shbail, M. O., Esra'a, B., Alshurafat, H., Ananzeh, H., & Al Kurdi, B. H. (2021).
Factors affecting online cheating by accounting students: the relevance of social
factors and the fraud triangle model factors. Academy of Strategic Management
Journal, 20, 1-16. https://doi.org/10.1016/j.dib.2021.107732
Amzalag, M., Shapira, N., & Dolev, N. (2021). Two sides of the coin: lack of academic
integrity in exams during the corona pandemic, students' and lecturers'
perceptions. Journal of Academic Ethics, 1-21.
Bagozzi, R. and Yi, Y. (1988) On the Evaluation of Structural Equation Models. Journal
of the Academy of Marketing Sciences, 16, 74-94.
http://dx.doi.org/10.1007/BF02723327
Baldwin, D. C., Daugherty, S. R., Rowley, B. D., & Schwartz, M. D. (1996). Cheating
in medical school: A survey of second-year students at 31 schools. Academic
Medicine 71, 267–273.
 69
Ballantine, J. A., Larres, P. M., & Mulgrew, M. (2014, March). Determinants of
academic cheating behavior: The future for accountancy in Ireland. In Accounting
forum (Vol. 38, No. 1, pp. 55-66). https://doi.org/10.1016/j.accfor.2013.08.002
Beck, L., & Ajzen, I. (1991). Predicting dishonest actions using the Theory of Planned
Behavior. Journal of Research in Personality, 25(3), 285–301.
Blankenship, K. L., & Whitley, B. E. (2000) Relation of general deviance to academic
dishonesty. Ethics and Behavior, 10(1), 1–12.
Carpenter, D. D., Harding, T. S., Finelli, C. J., Montgomery, S. M., & Passow, H. J.
(2006). Engineering students' perceptions of and attitudes towards cheating. Journal
of Engineering Education, 95(3), 181-194. http://dx.doi.org/10.1002/j.2168-
9830.2006.tb00891
Carpenter, D.D., Harding, T.S., Finelli, C.J. et al. Does academic dishonesty relate to
unethical behavior in professional practice? An exploratory study. SCI ENG ETHICS
10, 311–324 (2004). https://doi.org/10.1007/s11948-004-0027-3
Chapman, K., Davis, R., Toy, D., & Wright, L. (2004). Academic integrity in the
business school environment: I’ll get by with a little help from my friends. Journal
of Marketing Education, 26, 236–249. https://doi.org/10.1177/0273475304268779
Cheung, S. F., & Chan, D. K. (2000). The role of perceived behavioral control in
predicting human behavior: A meta-analytic review of studies on the theory of
planned behavior. Unpublished manuscript, Chinese University of Hong Kong, 1-
47.
Chin, W. W. (1998). The partial least squares approach for structural equation modeling.
https://psycnet.apa.org/record/1998-07269-010
Chudzicka-Czupała, A., Grabowski, D., Mello, A. L., Kuntz, J., Zaharia, D. V., Hapon,
N., ... & Börü, D. (2016). Application of the theory of planned behavior in academic
cheating research–cross-cultural comparison. Ethics & Behavior, 26(8), 638-659.
Chun-Hua Hsiao & Chyan Yang (2011) The Impact of Professional Unethical Beliefs
on Cheating Intention. Ethics & Behavior, 21:4, 301-316.
http://dx.doi.org/10.1080/10508422.2011.585597
DeVellis, R. (2012). Scale Development Theory and Applications. Sage Publications,
New York.
 70
DeVellis, R.F. and Thorpe, C.T. (2021fjalili) Scale Development: Theory and
Applications. Sage Publications, Thousand Oaks.
Devlin, M., & Gray, K. (2007). In their own words: a qualitative study of the reasons
Australian university students plagiarize. Higher Education Research &
Development, 26(2), 181–198. https://doi.org/10.1080/07294360701310805
Dordoy, A. (2002, July). Cheating and plagiarism: student and staff perceptions at
Northumbria. In Proceedings of the Northumbria Conference (Vol. 4).
Dzakadzie, Yayra. (2021). MODERATION EFFECT OF MORAL OBLIGATION ON
STUDENT’S INTENTION TOWARDS ACADEMIC DISHONEST
BEHAVIOUR: THE CASE OF PUBLIC UNIVERSITIES IN GHANA. Innovare
Journal of Education. 24-27. http://dx.doi.org/10.22159/ijoe.2021v9i6.42814
Fass, R. A. (1990). Cheating and plagiarism. In W. W. May (Ed.), Ethics in higher
education (pp. 170–184). New York: Macmillan Publishers.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: An
introduction to theory and research. Reading, MA: Addison-Wesley.
Fornell, C., & Larcker, D. F. (1981). Structural equation models with unobservable
variables and measurement error: Algebra and statistics. Journal of Marketing
Research, 18(3), 328–388. http://dx.doi.org/10.2307/3150980
Gaberson, K. B. (1997). Academic dishonesty among nursing students. Nursing Forum,
32(3), 14–20. https://doi.org/10.1111/j.1744-6198.1997.tb00205.x
Garrett, N., Lazzaro, S. C., Ariely, D., & Sharot, T. (2016). The brain adapts to
dishonesty. Nature Neuroscience, 19(12), 1727–1732.
https://doi.org/10.1038/nn.4426
Gullifer, J., & Tyson, G. A. (2010). Exploring university students' perceptions of
plagiarism: A focus group study. Studies in Higher Education, 35(4), 463-481.
http://dx.doi.org/10.1080/03075070903096508
Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: Indeed a Silver Bullet.
Journal of Marketing Theory and Practice, 19(2), 139–152.
https://doi.org/10.2753/MTP1069-6679190202
Hair, J. F., Risher, J. J., Sarstedt, M., & Ringle, C. M. (2019). When to use and how to
report the results of PLS-SEM. European Business Review, 31(1), 2–24.
https://doi.org/10.1108/EBR-11-2018-0203
 71
Hair, J. F., Sarstedt, M., Ringle, C. M., & Gudergan, S. P. (2018). Advanced issues in
partial least squares structural equation modeling (PLS-SEM). Thousand Oaks, CA:
Sage.
Hair, J., Hult, G. T. M., Ringle, C. M., & Sarstedt, M. (2022). A primer on partial least
squares structural equation modeling (PLS-SEM). SAGE Publications, Incorporated.
Hair, J.F., Hult, G.T.M., Ringle, C.M., Sarstedt, M., Danks, N.P., Ray, S. (2021). An
Introduction to Structural Equation Modeling. In: Partial Least Squares Structural
Equation Modeling (PLS-SEM) Using R. Classroom Companion: Business.
Springer, Cham. https://doi.org/10.1007/978-3-030-80519-7_1
Hair, J.F., Matthews, L.M., Matthews, R.L. and Sarstedt, M. (2017). PLS-SEM or
CBSEM: updated guidelines on which method to use. International Journal of
Multivariate Data Analysis, 1(2), 107-123.
https://doi.org/10.1504/IJMDA.2017.087624
Hair, J.F., Ringle, C.M. and Sarstedt, M. (2013) Partial Least Squares Structural
Equation Modeling: Rigorous Applications, Better Results and Higher Acceptance.
Long Range Planning, 46, 1-12. https://doi.org/10.1016/j.lrp.2013.01.001
Hall, P. A., & Fong, G. T. (2007). Temporal self-regulation theory: A model for
individual health behavior. Health Psychology Review, 1(1), 6–52.
https://doi.org/10.1080/17437190701492437
Harding, T. S., Mayhew, M. J., Finelli, C., & Carpenter, D. (2007). The theory of
planned behavior as a model of academic dishonesty in humanities and engineering
undergraduate. Ethic and Behaviour, 7, 1-23.
Hendy, N. T., & Montargot, N. (2019). Understanding Academic dishonesty among
business school students in France using the theory of planned behavior. The
International Journal of Management Education, 17(1), 85–93.
https://doi.org/10.1016/j.ijme.2018.12.003
Henseler, J., Ringle, C. M., & Sarstedt, M. (2015). A new criterion for assessing
discriminant validity in variance-based structural equation modeling. Journal of the
Academy of Marketing Science, 43(1), 115–135. https://doi.org/10.1007/s11747-
014-0403-8
 72
Henseler, J., Ringle, C. M., & Sinkovics, R. R. (2009). The use of partial least squares
path modeling in international marketing. Advances in International Marketing, 20,
277–319. https://doi.org/10.1108/s1474-7979(2009)0000020014
Hilbert, G. A. (1985). Involvement of nursing students in unethical classroom and
clinical behaviors. Journal of Professional Nursing, 1, 230–234.
Hoàng, T. T. H., Đặng, K. L., Nguyễn, T. H., Nguyễn, N. L., Tô, T. H., & Phạm, X. T.
(2024). Các nhân tố tác động đến “Liêm chính học thuật” của sinh viên đại học trong
bối cảnh phát triển trí tuệ nhân tạo. Tạp chí Giáo dục, 47-52.
Hoàng, T. T. H., Lê, P. H., Đào, T. M. L., & Nguyễn, T. N. (2023). Các yếu tố ảnh
hưởng đến “Liêm chính học thuật” của sinh viên trong bối cảnh chuyển đổi số tại
các trường đại học. Tạp chí Giáo dục, 23(12), 29-33.
Hofstede, G. (2001), Culture’s Consequences: Comparing Values, Behaviors,
Institutions, and Organizations Across Nations, 2nd ed. Sage, Thousand Oaks, CA.
https://doi.org/10.1016/S0005-7967(02)00184-5
Jalilian, F., Moazami, P., Mirzaei-Alavijeh, M., Moazami, A. M., & Jalili, C. (2016).
Sensation seeking and the intention to cheating among college students: An
application of the theory of planned behavior. Research Journal of Applied Sciences,
11(8), 645–649.
Jensen, L.; Arnett, J.; Feldman, S. and Cauffman, E. (2002). “It’s wrong but everybody
does it: Academic dishonesty among high school and college students”,
Contemporary Educational Psychology, 27, 209- 228.
Johnston, K. L., & White, K. M. (2003). Binge-drinking: A test of the role of group
norms in the theory of planned behaviour. Psychology and health, 18(1), 63-77.
Kam, C. C. S., Hue, M. T., & Cheung, H. Y. (2018). Academic dishonesty among Hong
Kong secondary school students: application of theory of planned
behaviour. Educational Psychology, 38(7), 945–963.
https://doi.org/10.1080/01443410.2018.1454588
Kasler, J., Hen, M., & Sharabi-Nov, A. (2019). Academic integrity in higher education:
The case of a medium-size college in the Galilee, Israel. Journal of Academic Ethics,
17(2), 151–167. https://doi.org/10.1007/s10805-018-9307-4
 73
Kendra Cherry, M. (2011, July 6). How can our attitudes change and influence
behaviors? Verywell Mind. https://www.verywellmind.com/attitudes-how-they-
form-change-shape-behavior-2795897
Kohlberg, L. (1976). Moral stages and moralization: The cognitive-developmental
approach. In T. Likona (Ed.), Moral development and behavior. New York: Holt,
Rinehart, & Winston.
La Barbera, F., & Ajzen, I. (2021). Moderating role of perceived behavioral control in
the theory of planned behavior: A preregistered study. Journal of Theoretical Social
Psychology, 5(1), 35-45.
Latimer, A. E., & Martin Ginis, K. A. (2005). The Theory of Planned Behavior in
prediction of leisure time physical activity among individuals with spinal cord
injury. Rehabilitation Psychology, 50(4), 389.
Latour, B. (2002). War of the worlds: What about peace? (C. Bigg, Trans.). Chicago:
Prickly Paradigm Press.
Lin, CH.S., Wen, LY.M. Academic dishonesty in higher education—a nationwide study
in Taiwan. High Educ 54, 85–97 (2007). https://doi.org/10.1007/s10734-006-9047-
z
Martin, D. E. (2012). Culture and unethical conduct: Understanding the impact of
individualism and collectivism on actual plagiarism. Management Learning, 43(3),
261–273. https://doi.org/10.1177/1350507611428119
Mayhew, M. J., Hubbard, S. M., Finelli, C. J., Harding, T. S., & Carpenter, D. D. (2009).
Using structural equation modeling to validate the theory of planned behavior as a
model for predicting student cheating. The Review of Higher Education, 32(4), 441-
468. https://doi.org/10.1353/rhe.0.0080
McCabe, D. L. (1992). The influence of situational students. Sociological Inquiry 62(3):
365-374. https://doi.org/10.1111/j.1475-682X.1992.tb00287.x
McCabe, D. L., & Trevino, L. K. (1993). Academic dishonesty: Honor codes and other
contextual influences. The Journal of Higher Education, 64(5), 522.
https://doi.org/10.2307/2959991
McCabe, D. L., & Trevino, L. K. (1996). What We Know About Cheating In College
Longitudinal Trends and Recent Developments. Change, 28(1), 28–33.
https://doi.org/10.1080/00091383.1996.10544253
 74
McCabe, D. L., & Trevino, L. K. (1997). Individual and contextual influences on
academic dishonesty: A multicampus investigation. Research in Higher
Education, 38(3), 379–396. http://www.jstor.org/stable/40196302
McCabe, D. L., Treviño, L. K., & Butterfield, K. D. (2001). Cheating in academic
institutions: A decade of research. Ethics &Behavior, 11(3), 219-232.
https://psycnet.apa.org/doi/10.1207/S15327019EB1103_2
McCabe, D. L., Trevino, L. K., & Butterfield, K. D. (2002). Honor codes and other
contextual influences on academic integrity: A replication and extension to modified
honor code settings. Research in Higher Education, 43, 357–378.
https://doi.org/10.1023/A:1014893102151
McCabe, D.L., Feghali, T. & Abdallah, H. Academic Dishonesty in the Middle East:
Individual and Contextual Factors. Res High Educ 49, 451–467 (2008).
https://doi.org/10.1007/s11162-008-9092-9
McCall Jr, M. W. (1988). Developing executives through work experiences. People and
Strategy, 11(1), 1.
McEachan, R. R. C., Conner, M., Taylor, N. J., & Lawton, R. J. (2011). Prospective
prediction of health-related behaviours with the Theory of Planned Behaviour: a
meta-analysis. Health Psychology Review, 5(2), 97–144.
https://doi.org/10.1080/17437199.2010.521684
Muñoz García A, Gil-Gómez de Liaño B, Pascual-Ezama D. The Role of Cognition in
Dishonest Behavior. Brain Sciences. 2023; 13(3):394.
https://doi.org/10.3390/brainsci13030394
Mustapha, R., Hussin, Z., Siraj, S., & Darusalam, G. (2016). Does Islamic religiosity
influence the cheating intention among Malaysian Muslim students? A modified
theory of planned behavior. International Journal of Academic Research in Business
and Social Sciences, 6(12), 389–406.
Nonis, S., & Swift, C. O. (2001). An examination of the relationship between academic
dishonesty and workplace dishonesty: A multicampus investigation. Journal of
Education for Business, 77(2), 69–77.
Nunnally, J.C. and Bernstein, I.H. (1994). The Assessment of Reliability. Psychometric
Theory, 3, 248-292.
 75
Ogilby, S. M. (1995). The ethics of academic behavior: Will it affect professional
behavior? Journal of Education for Business, 71(2), 92–96.
Passow, Honor J.; Mayhew, Matthew J.; Finelli, Cynthia J.; Harding, Trevor S.;
Carpenter, Donald D.; (2006). "Factors influencing engineering students’ decisions
to cheat by type of assessment." Research in Higher Education 47(6): 1-42.
http://hdl.handle.net/2027.42/42694
Pham, C. H., Vu, N. H., Le, A. Q., & Trinh, L. T. N. (2021). Students’ plagiarism
behavior: The case of universities in Vietnam. Elementary Education Online, 20(2),
341–351. https://ilkogretim-online.org/index.php/pub/article/view/1703
Plagiarism. (n.d.). Ox.ac.uk. Retrieved August 24, 2024, from
https://www.ox.ac.uk/students/academic/guidance/skills/plagiarism
Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common
Method Biases in Behavioral Research: A Critical Review of the Literature and
Recommended Remedies. Journal of Applied Psychology, 88, 879-903.
http://dx.doi.org/10.1037/0021-9010.88.5.879
Ramdhani, N. (2011). Preparation of Measuring Tools Based on the Theory of Planned
Behavior. Psychological Bulletin, 19(2), 55-69.
Salehi, M., & Gholampour, S. (2021). Cheating on exams: Investigating reasons,
attitudes, and the role of demographic variables. Sage Open, 11(2).
http://dx.doi.org/10.1177/21582440211004156
Schwarzer, R. (2001). Social-Cognitive Factors in Changing Health-Related
Behaviors. Current Directions in Psychological Science, 10(2), 47-
51. https://doi.org/10.1111/1467-8721.00112
Sims, R. L. (1993). The relationship between academic dishonesty and unethical
business practices. Journal of Education for Business, 68(4), 207–211.
Sims, R.R., Sims, S.J. Increasing applied business ethics courses in business school
curricula. J Bus Ethics 10, 211–219 (1991). https://doi.org/10.1007/BF00383158
Snyder, S. S., & Feldman, D. H. (1984). Phases of transition in cognitive development:
Evidence from the domain of spatial representation. Child Development, 55(3), 981–
989.
 76
Spraggon, J. Individual Decision Making in a Negative Externality
Experiment. Experimental Economics 7, 249–269 (2004).
https://doi.org/10.1023/B:EXEC.0000040560.94572.60
Stone, T. H., Jawahar, I. M., & Kisamore, J. L. (2009). Using the theory of planned
behavior and cheating justifications to predict academic misconduct. Career
Development International, 14(3), 221–241.
https://doi.org/10.1108/13620430910966415
Stone, T.; Kisamore, J. and Jawahar, I. (2007). “ Predicting academic dishonesty:
Theory of planned behavior and personality”, Proceedings of the 2007 Management
Education Division of the Administrative Sciences Association of Canada.
Thomas H. Stone, I.M. Jawahar, Jennifer L. Kisamore, (2009) "Using the theory of
planned behavior and cheating justifications to predict academic misconduct".
Career Development International, Vol. 14 Issue: 3, pp.221-241,
https://doi.org/10.1108/13620430910966415
Thomas, D. (2017). Factors that explain academic dishonesty among university students
in Thailand. Ethics & Behavior, 27(2), 140–154.
https://doi.org/10.1080/10508422.2015.1131160
Todd-Mancillas, W. R. (1987). Academic dishonesty among communication students
and professionals: Some consequences and what might be done about them. Paper
presented at the annual meeting of the Speech Communication Association, Boston,
MA.
Triandis, H. C., & Gelfand, M. J. (1998). Converging measurement of horizontal and
vertical individualism and collectivism. Journal of Personality and Social
Psychology, 74(1), 118–128. https://doi.org/10.1037/0022-3514.74.1.118
Uc, Tuan, D. T.-A., Giao, D.-A., & Vucong. (2021). Academic integrity: The case of
Vietnam. Psychology and Education, 58(5), 4028–4043.
http://psychologyandeducation.net/pae/index.php/pae/article/view/6192
Uzun, A. M., & Kilis, S. (2020). Investigating antecedents of plagiarism using extended
theory of planned behavior. Computers & Education, 144, 103700.
https://doi.org/10.1016/j.compedu.2019.103700
Vũ Đ. H., & Long N. T. (2021). Đánh giá liêm chính học thuật của sinh viên qua nhận
thức của sinh viên về môi trường học thuật và hành vi không trung thực học
 77
thuật. TẠP CHÍ KHOA HỌC ĐẠI HỌC MỞ THÀNH PHỐ HỒ CHÍ MINH - KINH
TẾ VÀ QUẢN TRỊ KINH DOANH, 16(1), 46–63.
https://doi.org/10.46223/HCMCOUJS.econ.vi.16.1.1389.1102
Whitley, B. E. (1998). Factors Associated with Cheating among College Students: A
Review. Research in Higher Education, 39, 235-274.
https://doi.org/10.1023/A:1018724900565
Whitley, B. E., & Keith-Spiegel, P. (2001). Academic Dishonesty. In Psychology Press
eBooks. https://doi.org/10.4324/9781410604279
Yusliza, M., Saputra, J., Fawehinmi, O., Mat, N., & Mohamed, M. (2020). The
mediating role of justification on the relationship of subjective norms, perceived
behavioral control and attitude on intention to cheat among students. Management
science letters, 10(16), 3767-3776.
Zhang, Y. Academic cheating as planned behavior: the effects of perceived behavioral
control and individualism-collectivism orientations. High Educ 87, 567–590 (2024).
https://doi.org/10.1007/s10734-023-01024-w
Zhang, Y., & Yin, H. (2020). Collaborative cheating among Chinese college students:
The effects of peer influence and individualism-collectivism orientations.
Assessment & Evaluation in Higher Education, 45(1), 54–69.
https://doi.org/10.1080/02602938.2019.1608504