Ncmaren Ib5clck45 Final-Report n06

UNIVERSITY OF ECONOMICS HO CHI MINH CITY
SCHOOL OF INTERNATIONAL BUSINESS - MARKETING
----------
FINAL REPORT
MARKETING RESEARCH
TITLE:
PRIVACY RISK AWARENESS AND INTENT TO DISLOSE
PERSONAL INFORMATION OF USERS USING TWO SOCIAL
NETWORKS:
FACEBOOK AND INSTAGRAM.
Supervior: PhD. DINH TIEN MINH

Class: DH45IBCO5
Group: 06
HO CHI MINH CITY, 2021.

FINAL REPORT
MARKETING RESEARCH
TITLE:
PRIVACY RISK AWARENESS AND INTENT TO DISLOSE
PERSONAL INFORMATION OF USERS USING TWO
SOCIAL NETWORKS: FACEBOOK AND INSTAGRAM.
SUPERVISOR: PHD. DINH TIEN MINH

CLASS: DH45IBCO5
GROUP: 06
MEMBERS:
NAME STUDENTS EMAIL

ID
Nguyễn Thị Ngọc 31191023185 [email protected]
Duyên
Nguyễn Bình 31191026089 [email protected]
Thảo Nhi
Puih Bé 31191026575 [email protected]
Bế Nguyễn Hằng 31191022375 [email protected]
Vũ Thanh Hiếu 31191023313 [email protected]
Table of Contents
THANK YOU............................................................................................................................1
PROJECT SUMMARY.............................................................................................................2
CHAPTER 1: INTRODUCTION..............................................................................................3
1. TOPIC:............................................................................................................................3
2. THE REASON TO CHOOSE THE TOPIC:..................................................................3
3. THE OBJECT OF STUDY:............................................................................................3
4. THE OBJECT OF SURVEY:.........................................................................................3
5. RESEARCH OBJECTIVES:..........................................................................................4
6. SCOPE OF THE STUDY:..............................................................................................4
7. RESEARCH METHODS:..............................................................................................4
8. SUBJECT LIMITED:.....................................................................................................5
CHAPTER 2: THEORETICAL FRAMEWORK................................................................6
1. RELATED THEORIES:....................................................................................................6
1.1. TPR – Theory of Perceived Risk (Bauer, 1960)..........................................................6
1.2. Theory of Planned Behavior (Ajzen, 1991).................................................................7
2. RELATED RESEARCH ARTICLES:...............................................................................8
2.1. Popular sources:...........................................................................................................8
2.2. Scholarly sources:......................................................................................................13
2.3. Government sources:.................................................................................................19
2.4. Commercial sources:..................................................................................................20
3. RESEARCH HYPOTHESIS:...........................................................................................21
3.1. Effect of trust in media on Facebook and Instagram user's privacy risk perception: 21
3.2. Effect of perceived benefits on Facebook and Instagram user's privacy risk
perception:........................................................................................................................22
3.3. Effect of habit on Facebook and Instagram user's privacy risk perception:..............23
3.4. Effect of personality on Facebook and Instagram user's privacy risk perception:....24
3.5. Effect of knowledge on Facebook and Instagram user's privacy risk perception:.....25
3.6. Effect of perception of privacy risk on personal information disclosure behavior:. .26
4. CONCLUSION THEORY FOR GROUP TOPIC:..........................................................27
CHAPTER 3: RESEARCH DESIGN..................................................................................29
1. RESEARCH PROCESS:..............................................................................................29
2. QUALITATIVE RESEARCH:.....................................................................................30
2.1. How to do it:..............................................................................................................30
2.2. Results of qualitative research...................................................................................31
2.3. Questionnaire design..................................................................................................42
3. QUANTITATIVE RESEARCH:.....................................................................................43
3.1. Sampling method.......................................................................................................43
3.2. Sample size................................................................................................................43
3.3. Processing and analyzing data...................................................................................44
CHAPTER 4: QUANTITATIVE RESEARCH RESULTS...............................................48
1. SAMPLE DESCRIPTION:...........................................................................................48
1.1. Sample structure by sex.............................................................................................48
1.2. Sample structure by age group...................................................................................48
1.3. Sample structure by industry group:..........................................................................49
1.4. Sample structure according to usage habits:..............................................................50
1.5. Sample structure according to frequency of use:.......................................................51
1.6. Sample structure by number of friends:.....................................................................51
1.7. Sample structure by time of use:................................................................................52
2. CHECK THE SCALE:..................................................................................................53
2.1. Testing the scale of “Belief”......................................................................................54
2.2. Testing the scale of “Perception of benefits”.............................................................54
2.3. Testing the scale of “Habit”.......................................................................................55
2.4. Testing the scale of the “Personalization”.................................................................56
2.5. Testing the scale of “Knowledge”.............................................................................56
2.6. Testing the scale of “Privacy Invasions Experience”................................................57
2.7. Testing the scale of "Privacy Policy".........................................................................58
2.8. Testing the intermediate variable scale “Privacy risk perception”...........................58
2.9. Testing the dependent variable scale “Intent to disclose personal information”.......59
3. EXPLORATORY FACTOR ANALYSIS (EFA):........................................................61
3.1. Exploratory factor analysis with independent variables............................................61
3.3. Exploratory factor analysis before running CFA and SEM:......................................66
4. CONFIRMATORY FACTOR ANALYSIS (CFA):.....................................................67
5. STRUCTURAL EQUATION MODELING (SEM):....................................................69
6. TESTING THE RESEARCH HYPOTHESIS:.............................................................73
CHAPTER 5: CONCLUSIONS............................................................................................75
1. SUMMARY RESULTS:...............................................................................................75
2. MEANING:...................................................................................................................75
3. LIMITATION OF THEME AND FUTURE RESEARCH DIRECTIONS.................79
4. SOLUTION...................................................................................................................80
REFERENCES.......................................................................................................................83
APPENDIX A: QUALIFICATION QUESTION LIST FOR SPECIALISTS AND FOCUS
GROUPS..............................................................................................................................87
APPENDIX B: QUALITATIVE RESPONSE LIST...........................................................88
APPENDIX C: QUANTITATIVE QUESTIONNAIRE FOR SURVEY............................89
APPENDIX D: CFA DIAGRAM........................................................................................98
APPENDIX E: SECONDARY DATA SOURCES.............................................................99
PERFORMANCE ASSESSMENT TABLE:.....................................................................100
TABLE OF PICTURES
Figure 1. Model of reference for research…………………………………….8
Figure 3.Model of reference for research……………………………………....10
Figure 4. Model of reference for research……………………………………...14
Figure 5. Model of reference for research……………………………………..15
Figure 9. Model of reference for research……………………………………...20
Figure 11. Research model…………………………………………………….28
Figure 12. Research model after qualitative research………………………….33
Figure 13. Results of Model and Hypotheses Testing........................................70
TABLE OF BOARD
Table 1: Results of qualitative research and scale……………………………..38
Table 2: Calculate the number of survey samples……………………………..44
Table 3: User's gender………………………………………………………….48
Table 4: User's age……………………………………………………………..49
Table 5: User's occupation……………………………………………………..50
Table 6: Using Facebook and Instagram……………………………………….51
Table 7: Frequency of Use Facebook and Instagram…………………………..51
Table 8: a user's number of Facebook and Instagram friends………………….52
Table 9: Used time Facebook and Instagram…………………………………..53
Table 10: Reliability Statistics and Item-Total Statistics………………………54
Table 19: The final summary table of the scale test……………………………60
Table 20: 1st- Rotated Component Matrix……………………………………..62
Table 21: 2nd - Rotated Component Matrix…………………………………....63
Table 22: 1st-rotation matrix…………………………………...........................64
Table 23: 2nd-rotation matrix…………………………………..........................65
Table 24: Summary for results KMO……………………….………................65
Table 25: Pattern Matrix……………………….………....................................66
Table 26: CMIN……………………….……….................................................68
Table 27: RMR, GFI……………………….………..........................................68
Table 28: Baseline Comparisons…………….………........................................69
Table 29: RMSEA……………………….………..............................................69
Table 30: Summary of the normalized and values……......................................71
Table 31: Indirect and direct effects of variables…………...…………………72
THANK YOU
First of all, our group would like to thank the teachers and friends who have
encouraged and helped us a lot in the process of conducting this scientific
research.
In particular, the team would like to express their deepest gratitude to Mr. Dinh
Tien Minh and Ms. Linh, who have taught and guided the group wholeheartedly
throughout the process of conducting and completing this research paper.
In addition, the team would also like to thank the teachers and students named
below who helped and supported us in providing orientation, suggestions and
additional information to complete the research paper throughout last time.
- Master Hoang Cuu Long
- Mr. Tran Thanh Tung
- Le Hoang An
- Nguyen Thao Nguyen
- Nguyen Dang Minh Quan
- Ro Cham Bien
- Nguyen Thi Anh Thu
- Tran Nguyen Nhu Quynh
- Mr. Tran Anh Minh
And the friends who participated in the survey to help my group.
Our team sincerely thank you!
1
PROJECT SUMMARY
Today, with the rapid development of 4.0 technology, more and more people
are moving to a second life: the digital world. And everyone, in one way or
another, is facing information privacy concerns, since engaging in sharing and
communicating on social networking sites entails saving money. disclose
personal information. Although today there are threats such as privacy issues,
malicious use of information, invasion of information access rights, user safety
risks, and negative effects of social networks, ... people still continue to reveal a
large amount of personal information on online social networks, most notably
the two major social networking sites, Facebook and Instagram.
One might ask: do Facebook and Instagram users really take their privacy
seriously? And how does this affect their perception of privacy risks as a result
of the personal information they disclose? And what factors exist that can
influence the intention to disclose personal information on Facebook and
Instagram?
Facebook and Instagram are constantly and rapidly updated (options for
personal sharing of information through digital media) and the technologies that
enable individuals to use the service require the constant attention of
researchers. The main objective of the study is to better understand the factors
that motivate online social network users to disclose personal information in
order to analyze and assess users' perceptions of privacy risksdisclose personal
information through two social networking sites, Facebook and Instagram.
“ Nobody is literally forced to join an online social network, and most

networks we know about encourage, but do not force users to reveal personal
information. And yet, one cannot help but marvel at the nature, amount, and
detail of the personal information some users provide, and ponder how
informed this information sharing is.”
—Acquisti & Gross (2006, p. 37)
2
CHAPTER 1: INTRODUCTION
1. TOPIC:
Privacy risk awareness and intent to disclose personal information of users
using two social networks: Facebook and Instagram.
2. THE REASON TO CHOOSE THE TOPIC:

With the advancement of technology 4.0, the Internet has become an integral
part of our everyday lives. Online social networking sites arose as a result of the
Internet's arrival and growing strongers. People use social networking sites to
keep up with news, chat with friends, strengthen relationships, and even conduct
business and earn extra money. The monitoring of security and safety for users
is becoming more difficult as the number of people using social networking
sites, especially the two main social networking sites Facebook and Instagram,
grows.
As a result of the above practice, our group debated and agreed on the following
topic: "Being aware of privacy risks and the intention to reveal personal details
of Facebook and Instagram users." Our team will learn more about the factors
that influence users' perceptions of privacy risks on two major social
networking sites, Facebook and Instagram, and how that understanding
influences their willingness to share personal information in this research paper.
It will then assist users in gaining a greater understanding of privacy on social
networking sites so that they can secure themselves on these sites.
3. THE OBJECT OF STUDY:

User awareness regarding privacy risks and disclosure of personal information
on two social networks: Facebook and Instagram.
4. THE OBJECT OF SURVEY:

Users of social networking sites Facebook and Instagram in Vietnam belong to
the following age groups:
- From 15 to 19
- From 20 to 24
- From 25 to 29
3
5. RESEARCH OBJECTIVES:
Research on the relationship of privacy awareness with the intent to disclose
personal information.
Discover factors affecting risk perception as well as intent to disclose personal
information of Facebook and Instagram users.
Propose solutions to improve risk awareness and publicize personal information
of users of social networks Facebook and Instagram.
6. SCOPE OF THE STUDY:

People aged 15 to 29 in Vietnam have been using two online social networking
sites Facebook and Instagram.
7. RESEARCH METHODS:
The study was carried out in two main steps:
(1) Preliminary research using a qualitative research method.
(2) Formal research to collect data for quantitative research.
- Qualitative research: is used to explore nature, directly explore ideas and to
provide a preliminary description of the research model of the group. From the
results of qualitative research combined with previous research articles related
to the topic, make amendments and supplements to the research model; at the
same time, develop a formal questionnaire to match the meaning of the scales
and the subjects for sampling. Conduct qualitative research: Face-to-face
interviews or messaging on online social networking sites. The study population
was 8 people. Includes 1 expert- with deep understanding of social networking,
and a focus group of 7 people. Focus on understanding the factors affecting the
perception of privacy risks, and direct respondents to the factors used in the
research model, scale construction, questions, concepts, terminology related.
- Quantitative research: used to collect data used in the research paper and the
final results. Conduct quantitative research: Create an online questionnaire, and
send it to research subjects to answer and collect data for the research paper.
4
8. SUBJECT LIMITED:
The study focused only on people between the ages of 15 and 29. Since,
according to our research, this target audience uses the social media platforms
Facebook and Instagram more often than other age groups.
5
CHAPTER 2: THEORETICAL FRAMEWORK
1. RELATED THEORIES:
1.1. TPR – Theory of Perceived Risk (Bauer, 1960)

According to Bauer (1960): Perceived risk is defined as consisting of two main
components: the probability of a loss and the subjective feeling of a bad
outcome.
The theory of risk perception TPR developed by Bauer states that the behavior
of consuming technology products Perceived risk information includes two
elements: (1) Perceived Risk with Product/Service (PRP) and (2) Perceived
Risk associated with Product/Service Perceived Risk in the Context of
OnlineTransaction (PRT).
* Risk perception component related to product/service: Types of perceived
risk: loss of functionality, loss of finance, loss of time, loss of opportunity, and
overall risk perception with the product/service (sum of the consumer's
perception of uncertainty or anxiety when purchasing the product).
* Risk perception components related to online trading: The risks that may
occur when consumers conduct e-commerce transactions on electronic means
and devices are related to: confidentiality, security authentication,
nonrepudiation, and overall risk perception of online transactions.
Risk perception theory (TPR)

(Source: Bauer, 1960)
6
1.2. Theory of Planned Behavior (Ajzen, 1991)
- The theory of intended behavior (TPB) (Ajzen, 1991), developed from the
theory of rational action (TRA; Ajzen & Fishbein, 1975), assumes that a
behavior can be predicted or explained by behavioral propensity to perform that
behavior. Behavioral tendencies are assumed to include motivational factors
that influence behavior, and are defined as the degree of effort with which
people attempt to perform. perform the behavior (Ajzen, 1991).
- Behavioral propensity is again a function of three factors. First, attitudes are
conceptualized as positive or negative evaluations of performance behavior. The
second factor is social influence which refers to the perceived social pressure to
perform or not to perform the behavior. Finally, the theory of planned behavior
TPB (Theory of Planned Behavior) was built by Ajzen by adding the perceived
behavioral control factor to the TRA model.
- The perceived behavioral control component reflects the ease or difficulty of
performing the behavior; This depends on the availability of resources and
opportunities to perform the behavior. Ajzen suggested that behavioral control
factors directly affect the tendency to perform the behavior, and that if the
person is accurate in his perception of the degree of control, then behavioral
control also predicts the behavior.
7
2. RELATED RESEARCH ARTICLES:
2.1. Popular sources:
2.1.1. Books:
*Consumers' ethical perceptions of social media analytics practices:
Risks, benefits and potential outcomes
-Nina Michael Idoua, Milena Maticevski-
- In this study, we model outcomes of the ethical perceptions of SMA, and of
trustworthiness, looking also at the roles of perceived risk in sharing
information on social media, and benefits.
- The research model:
Figure 1. Model of reference for research

- The results show that the impact of ethical perceptions and trustworthiness on
consumers' willingness to reveal information and the likelihood to take action
against organizations is mediated by the level of perceived risk in sharing
information on social media.
*Voluntary sharing and mandatory provision: Private information

disclosure on social networking sites
-Kai Lia, Liangqi Chenga, Ching-I Teng-
- This study divides private information disclosure into two modes: voluntary
sharing and mandatory provision. Based on the CPM theory, the selected
8
samples were concentrated among 21 to 30 year olds, this paper studied the
impact of various factors on disclosure willingness, enriching the existing
literature on disclosure.
- The results of this study:

(1) Perceived benefits have a stronger influence on the willingness of users to
disclose private information and are the main drivers of disclosure.
(2) Users who are more cautious about disclosure are more reluctant to disclose
private information to the website. Mandatory provision of private information
requires higher security from the network.
2.1.2. Newspapers:
*Understanding third-person perception about Internet privacy risks
- Hongliang Chen, David Atkin-
- This study aims to examine the effect of third parties in Internet privacy risk
perception. Support is found for a TPE model that suggests that users report
Internet privacy risk awareness to others rather than themselves, based on a
sample from Amazon MTurk.
9
- Research results indicate that third-party perception is one of the major
barriers preventing privacy protection from being applied. Another consequence
of a privacy breach involves a relationship conflict. Unauthorized access to
personal accounts can lead to inappropriate comments and posts that can cause
conflicts with online contacts, damage users' reputation and lose muscle mass.
However, some other users simply believe they won't experience negative
privacy experiences, while others are more likely to be victims.
- Drawn from TPE theory, this study develops a three-phase model that includes
five precursors of TPE awareness of Internet privacy risks - social distance,
privacy risk on Unidentifiable Internet, previous negative privacy experiences,
online privacy and consumption security awareness of online sources about
Internet privacy threats; TPE's awareness of Internet privacy risks; and the act
of intent to self-take privacy protections and refer it to others.
Figure 3.Model of reference for research
10
*“Who should I grant access to my post?”: Identifying the most suitable
privacy decisions on online social networks
- Jose Alemany, Elena Del Val, Ana Maria Garcia- Fornes-
- This study focuses on all the elements of communication in the information
disclosure process, taking into account the attributes of the message and the
recipient, and the impact on the value of the user benefit. Which refers to the
risk of personal information disclosure and the right to access information of
groups of people.
- Research has been done to understand how users evaluate the cost benefit of
online social disclosure for the different types of receivers identified and the
sensitivity of the message. Data from 400 respondents were collected and
analyzed using a partial least squares model.
- The findings of this study demonstrate a balanced variance between perceived
cost benefits and disclosure of sensitive information with different types of
receivers. Disclosing personal information to trusted recipients, influential
recipients, and recipients from peer circles has had a significant positive impact
on building social capital. Conversely, disclosing personal information to
recipients from family circles or unidentified recipients has a significant
negative impact on social capital building and even a significant positive effect.
to privacy concerns.
*A measure of the implicit value of privacy under risk.
-Alisa Frik, Alexia Gaudeul -
- In this study, the authors allowed individuals to play the privacy lottery and
took a measure of the value of privacy under risk (VPR) and experimentally
tested the validity of the measure. This was in a lab trial with 148 participants.
Individuals are asked to make an incentive series of decisions about whether to
take the risk of disclosing personal information to other participants.
- The results confirm that privacy risk willingness is driven by a complex range
of factors, including risk aversion, self-reported value of personal information,
and general attitudes. for privacy (derived from surveys). VPR does not depend
on whether there is a pre-existing threat to privacy. The authors found
qualifying support for the existence of an order effect, whereby presenting
financial options before privacy options resulted in less privacy concerns.
11
2.1.3. Blogs:
*Protecting the rights to personal data and information in the current social
networking environment in Vietnam
- On social networks in Vietnam such as facebook, instagram,...in recent years,
there have been many acts of revealing personal secrets, family secrets ...
beyond the will of individuals and families being exposed. . Many people have
not understood, misunderstood or intentionally did not understand the right to
freedom of expression, freedom of expression of personal opinions, freedom of
the press ... so accidentally or intentionally revealing privacy, personal secrets.
personalities, family secrets of other people, causing mixed reactions on social
networks.
=> Through analyzing the current situation of personal data protection in
Vietnam, the author proposes a number of solutions to contribute to improving
and improving the effectiveness of the enforcement of the law on personal data
protection on the environment. social networking school in Vietnam.
2.1.4. Social Media:

*The New Rules on Social Media, Privacy and Data Protection
- With the rapid growth of social networking sites such as facebook and
instagram, users are less cautious about protecting personal information so
concerns about privacy laws are also limited. . Only when the risks of personal
information leak, which greatly affects private life, do the issue of privacy
protection really come into play.
- This article covered new social media, privacy and data protection rules that
improve your security and significantly reduce the risks you and your customers
face.
2.1.5. Magazines:
*Don't neglect the privacy of your personal information
- In the era of globalization, connecting and interacting on social networks are
increasingly popular. However, very few Internet users in Vietnam care about
the security of personal information. The ease of many people when posting
12
personal information online makes their information more at risk of
exploitation.
- The article outlined the easy habits of users in disclosing personal information
through their social networking habits and outlined the consequences of losing
*Disclosure of personal information on the Internet:

- With the use of smart devices, artificial intelligence, but not all users
understand well about what they are using. Users do not envision the dangers
right next to them, from using the Internet in public places, which makes it seem
that smart mobile devices are "monitoring" their activities. These are the cases
of personal information disclosure that users often ignore, do not understand
why and do not know how to prevent.
- The article has shown easy ways to steal the personal information of users of
technology equipment, the habit of using social networking sites to commit bad
behaviors.
2.2. Scholarly sources:
*Understanding the Key Antecedents of Users’ Disclosing Behaviors on Social

Networking Sites: The Privacy Paradox
-Byoungsoo Kim and Daekil Kim-
- This study explored the formation mechanisms of users’ disclosing behaviors
from the perspectives of the privacy paradox.
- The findings show that users’ intention to disclose personal information has a
marginally significant effect on users’ disclosing behaviors. Moreover,
perceived control over personal information mitigates the level of privacy
concerns.
13
- Results: Consistent with expectations, users’ intention to disclose personal

information has a marginally significant effect on users’ disclosing behaviors,
thus supporting H1. Perceived benefits have a significant influence on both
users’ intention to disclose personal information and users’ disclosing
behaviors, thus supporting H2a and H2b. Trust in an SNS provider has a
significant influence on users’ intention to disclose personal information while
it has no significant effect on users’ disclosing intention. Perceived control over
personal information is positively associated with trust in SNS providers, users’
intention to disclose personal information, and users’ disclosing behaviors.
*Examining Self-Disclosure on Social Networking Sites: A Flow Theory and

Privacy Perspective
-George Oppong Appiagyei Ampong, Aseda Mensah, Adolph Sedem Yaw Adu,
John Agyekum Addae, Osaretin Kayode Omoregie and Kwame Simpe Ofori-
- Social media and other web 2.0 tools have provided users with the platform to
interact with and also disclose personal information to not only their friends and
acquaintances but also relative strangers with unprecedented ease. Interaction
and perceived control were found to have significant effects on self-disclosure.
14
- Results: Data collected from the survey was analyzed using the Partial Least
Square approach to Structural Equation Modelling performed on SmartPLS
Version 3 . Structural Equation Modelling allowed the researchers to test
relationships between latent variables in the proposed research model. To
validate the measurement model, we examined reliability, convergent validity
and discriminant validity. The reliability of the constructs was assessed with
Cronbach’s alpha and composite reliability.
*Antecedents of Self-Disclosure on Social Networking Sites (SNSs): A Study of

Facebook Users
-Ashraf Sharif, Saira Hanif Soroya, Shakil Ahmad and Khalid Mahmood-
- Self-disclosure on social networking sites leads to social capital development,
connectedness, and relationship building. The current study investigates the
antecedents of self-disclosure under the lens of the technology acceptance
model . The partial least squares-structural equation model analysis technique
was employed to analyze the data.
- Results: The study concludes that self-disclosure behavior is dependent on
multiple factors, individuals’ perceptions about the usefulness of personal
information sharing, trust in the community to whom personal information is
shared, trust in the medium. Facebook will keep its promises and will use
personal information fairly. Although individuals’ perceived ease of use does
15
not impact self-disclosure directly, it does, however, help them to understand
the potential benefits of self-disclosure.
*A survey of social media users privacy settings & information disclosure

-Mashael Aljohani, Alastair Nisbet, Kelly Blincoe-
- The aim of this research is to shed light on SNS user’s personal information
disclosure behaviours, their privacy protection settings, privacy policies and
users SNS privacy knowledge and awareness. The study was conducted to
identify the effect of gender, education status, and age on the degree of personal
information disclosure and protective privacy settings applied by the user, using
factor analysis.
- Research Results: This selection of results shows that many factors comprise
the profile decisions of users and those who choose to join as members. The
publicity over the risks of disclosing private information that may be used to
construct fake profiles, stalking and other nefarious activity seems to have had
little effect on many SNS users. The desire to be part of a community, often
with hundreds of friends, most of which the person will never meet and who
themselves may be using fake identities, seems to have only a modest effect on
the users’ sense of caution.
*Personal Information Disclosure on Online Social Networks

-Ruud H.G. Koehorst-
- This report discusses the results of an empirical study on the predictors of an
individuals’ intention to disclose personal information on online social networks
(OSNs). Beldad, De Jong, and Steehouder (2011) provide a solid basis with
their theoretical framework for information-related behaviors on the internet. In
their literature research, Beldad et al. mention the influence of benefits, trust,
risk perception, and habits on personal information disclosure or protection
behavior. In addition, the authors note the role of privacy concerns on risk
perception, and the influence privacy assurances and security features have on
trust.
- Data showed that the respondents’ (n = 491) habits were the strongest
significant predictor for one’s intention to disclose personal information on
OSNs. Subsequently, there is a significant influence of the benefits of sharing
personal information and the perceived control over this personal information
16
on the intention to disclose personal information on OSNs. In addition, it was
found that there was a causal relationship between both the trusting beliefs
concerning disclosing personal information and the perceived control over this
information, and the respondents’ privacy valuation.
*Information privacy disclosure on social network sites

- Purpose – As social network sites (SNS) have increasingly become one of the
most important channels for communication, the related privacy issues gain
more and more attention in both industry and academic research fields. This
study aims to connect the antecedents of information privacy disclosure on
SNS.
- Design/methodology/approach – Based on exchange theory, this study tries to
investigate the decision-making process for information privacy disclosure on
SNS. Factors from both user’s and website’s perspectives are taken into account
in the proposed model.
- The results suggest that an individual’s perceived benefits will increase their
willingness to disclose information privacy on SNS, but perceived risks
decrease this kind of willingness. The authors also find social network size,
personal innovativeness and incentive provision positively affect people’s
perceived benefits.
17
*Information privacy in e-service: Effect of organizational privacy assurances

on individual privacy concerns, perceptions, trust and self-disclosure behavior
- The objective of this study was to develop and validate a model that examined
the effect of organizational privacy assurances to users' privacy concerns,
perceptions, trusting beliefs and non-self-disclosure behavior.
- The study defines a consolidated model that addresses a number of constructs
related to organizational privacy assurances and outcomes that have only been
considered separately in previous studies. Earlier studies have identified some
variables and relationships that may explain perceptions, concerns, and behavior
associated with organizational privacy assurances but have not integrated them
into one comprehensive model. Earlier efforts to explain the organizational
privacy assurance mechanisms appear fragmented.
- This study provides a model that comprehensively explains the links between
organizational privacy-assurance-mechanisms-related constructs, privacy
concerns, perceptions, and trust and non-self-disclosure behavior. The model
identifies how individuals conceive the organizational self-policing level and
18
compares that with their own inherent need for privacy protection. The model
contributes to the theoretical development of privacy protection.
2.3. Government sources:
*Privacy Impact Assessment - OPA - Social Media Services

- Analysis performed in this Privacy Impact Assessment (PIA) focuses on
Personally Identifiable Information (PII) collected via Department of Labor
(DOL) branded functionality within Social Media Services such as Facebook
and YouTube ( collectively referred to as "Social Media Services")
DOL has implemented the following safeguards within the Department to assist
with the management of the Department's Social Media branded functionality:
+DOL Social Media Handbook (DLMS 5-600) - outlines processes for the
dissemination of information to the public when using Social Media Services
+Designated resources (DOL Social Media Coordinator) to maintain the
department's presence within these Social Media Services.
+Provided the DOL Social Media Coordinator with annual security awareness
and privacy training.
+DOL privacy policy on DOL.gov provides disclaimers on the use of Social
Media Services.
19
2.4. Commercial sources:
*Disclosing personal information to social networking site providers The role

of trust, risk and perceived benefits
- This thesis is set to investigate the factors that affect the intention to disclose
personal information to SNS providers.
- Through the questionnaire, a proposed framework was tested in order to
identify factors affecting peoples’ limitations of this study is that it suffers
sampling restrictions due to its sampling of only Swedish inhabitants.
*Examining users’ personal information sharing awareness, habits, and

practices in social networking sites and e-learning systems
- Reports of identity theft continue to be widely reported, while users continue
to share an increasing amount of personal information online, especially within
social networking sites and e-learning systems . However, the relationship
between habit and practice is not always clear. Habit theory has been validated
across many disciplines, with very limited attention in Information Systems. A
total of 390 responses were received, and path analysis was conducted to test
the hypotheses.
20
*Online Social Network - A Threat to Privacy and Security of Human Society

With over 1 billion users connected through online social media, user
confidentiality is becoming even more important and is widely argued in the
media and researched in academia. Social networking sites are a powerful and
fun way to communicate with the world. Social media is a good service because
it lets you share what you actually want to share, but it can also be used for
negative purposes, and in both cases you are responsible for your security. In
this paper we provide a brief overview of some attire to users’ privacy.
3. RESEARCH HYPOTHESIS:
3.1. Effect of trust in media on Facebook and Instagram user's privacy risk
perception:
Level of trust, this concept refers to the degree to which an organization is
deemed trustworthy and benevolent by consumers and possesses integrity as
well as essential skills and competencies (Caldwell & Clapham, 2003; Mayer &
Davis, 1999). According to a research paper by (Nina Michaelidou, Milena
Micevski) on Consumer Ethical Perceptions of Social Media Analysis Methods:
Potential Risks, Benefits, and Outcomes, states that: if the The lower the trust
level, the higher awareness of disregard for information disclosure on social
networks. The reliability of the product is low, the user is less likely to share
information in the media and the user tends to be against the manufacturer. It is
like the social exchange theory (SET) commonly used to form the concept of
trust, focusing on the rules of exchange, in which the interactions between one
party are conditional and dependent on the actions of the party. others
(Cropanzano & Mitchell, 2005). Consequently, confidence levels are assumed
21
to be reciprocal (eg, Croson & Buchan, 1999). Through the paper of (Nina
Michaelidou, Milena Micevski) client beliefs are informed by the theory of
justice, even in cases where they are not clearly specified and when there is no
clear interaction between the two. Under this theory, credibility is underpinned
by procedural justice, which requires the participation of or representation of
key stakeholders in the organization's processes. Likewise, the responsibility to
inform stakeholders, such as social media users, about the organization's goals
and the procedures used by the organization, and to ensure information
protection as part of activities. If an organization fails to complete a task, the
example does not provide information about the user's SMA (social network
analytics) activities related to the media collection of user information and data.
In society where users do not want this can lead to uncertainty and concern
about the consequences of sharing information on social networks, the problem
is the perception of privacy risks. In addition, lack of trustworthiness affects a
consumer's decision to provide fake information, such as an act of protection or
retaliation, expected by the consumer and trusting organizations to protect their
information (Punj, 2017). Such breaches of trustworthiness can also lead to
consumer action against a service provider in the form of a complaint.
*Hypothesize:
H1: High user trust in media has a positive impact on Facebook and Instagram
users' perceptions of privacy risk.
3.2. Effect of perceived benefits on Facebook and Instagram user's privacy

risk perception:
Literature on the benefits of disclosure often conceptualizes the benefits in a
“risk versus reward” calculation. This might be the result of the Social
Exchange Theory (which is often seen as the theoretical foundation of personal
information disclosure), that states that interpersonal relationships are based on
a subjective evaluation of benefits and costs (Homans, 1958, p. 606). The
Privacy Calculus Theory argues that some users feel that the returns for
disclosure offset the risk of their privacy being compromised (e.g. Dinev &
Hart, 2006; Culnan & Armstrong, 1999). Research found that people are willing
to sacrifice the safety of their personal information if the perceived benefits
outweighs the costs (for an overview, see Beldad et al, 2011, p. 225), and
despite concerns about privacy, adolescents are particularly receptive to the
22
potential benefits of disclosing personal information (Christofides, Muise &
Desmarais, 2009, p. 342).
Benefits that are associated with disclosure are plentiful: enjoyment (e.g.
Krasnova et al., 2009); self-presentation (e.g. Boyd, 2009); the ability to
maintain social ties (e.g. Ellison, Steinfield & Lampe, 2007)
*Hypothesize:
H2: Perceived benefits had a positive effect on intention to disclose personal
information on Facebook and Instagram.
3.3. Effect of habit on Facebook and Instagram user's privacy risk

perception:
Habit is characterized as a repetitive behavior that occurs as a result of the
automatic processing of stimulus signals rather than deliberate processing.We
can operate habits such as regularly disclosing personal information without the
need for deliberate processing and instead result from the automated processing
of excitation signals because it is linked to the habit of disclosing personal
information.
Habits are also an important factor affecting users' perception of privacy risks,
thereby influencing the intention to disclose personal information. In the
research articles that we found, there are 2 important articles that mention the
"Habit" factor that greatly affects the perception of users.
In the first study “Personal Information Sharing Habits ”, Habit has been
found to impact behavior beyond other factors (BurtonJones & Hubona, 2006),
and has been found to be a stronger predictor of behavior than intention (de
Bruijn, Kroeze, Oenema, & Brug, 2008; Kremers & Brug, 2008, Limayem et
al., 2007; Polites & Karahanna, 2012). For example, in one study habits were
shown to impact not only users’ intention to use IS, but also the intention to
continue to use IS.
The second research paper has shown that Habits apply well to OSN usage
behavior and habits can trigger intention automatically.
According to Lankton, McKnight, and Thatcher (2012), OSNs' conduct is
influenced by their behaviors (Here we can replace them with other social
networks like Facebook or Instagram). According to the findings of Ellison,
Steinfield, and Lampe (2007), questions about use patterns for Facebook users
mostly addressed the above definition.
23
The relationship between habit and continuation purpose is explained using
habit theory (Lankton, McKnight, and Thatcher (2012). Intention and users can
be automatically triggered by habit (Ajzan, 2002, p. 119). can elicit strong
feelings of friendliness against such behaviors based on previous habitual
practices, thus increasing the desire to continue the activity (Ellison, Steinfield
& Lampe, 2007 Beldad et al (2011) stated that the value of information
disclosure comes not just from the reasons people disclose information, but also
from the 'taste' of the disclosure itself (p. 226). Strater and Richter's results
support the possibility of strong effects of usage patterns (2007). They
discovered that some of their respondents were unsure why. What motivates
them to exchange information? Others, simply because they were used to filling
out paperwork, did not think twice about providing personal information when
asked.
*Hypothesize
H3: Intention to disclose personal information on Facebook and Instagram has
increased as a result of habit.
3.4. Effect of personality on Facebook and Instagram user's privacy risk

perception:
Personalizationis the use of information that businesses collect and analyze
customers, to create marketing activities and increase the relevancy of ads to
target, collect and analytics can be understood simply as basic or more specific
demographic information such as preferences, buying intent, and behavioral
patterns. According to the article by author Kai Li, Liangqi Cheng, Ching-I
Teng writes about "Voluntary and mandatory provisioning: Personal
information disclosed on social networking sites" claiming personalization
impacts positively come to social media users' risk perception. Research shows
that personalized services have significant security implications because the
premise of those services is the collection of large amounts of personal data
(Kobsa, 2007). An individual can provide important information such as name
and contact information to a website. If this is used improperly, it presents a
personal security risk and could lead to detection and tracking of consumers'
identities and behavior. These disclosures may lead to unwanted marketing,
advertising, price discrimination and unauthorized access (McKnight,
Choudhury & Kacmar, 2002). The collection of user information adds value to
the service, but consumers are concerned that their personal information being
24
collected and used to achieve personalization is at risk. privacy (Altman &
Taylor, 1973), i.e. risk awareness of user privacy. That is why we want to
hypothesize that personalization affects Facebook and Instagram user's privacy
perceptions just as it positively affects risk perception of social media users in
our research paper. by Kai Li, Liangqi Cheng, and Ching-I Teng.
*Hypothesize:
H4: Personalization positively affects Facebook and Instagram users'
perception of privacy risks.
3.5. Effect of knowledge on Facebook and Instagram user's privacy risk

perception:
According to the article by two authors (Hongliang Chen, David Atkin),
"Understanding third-party perceptions of privacy risks on the Internet". Having
mentioned an Internet privacy risk awareness factor positively predicts TPE's
perception of privacy risks on the Internet. Based on this research paper we
want to develop the issue of how understanding of social media users Facebook
and Instagram impacts perceptions of privacy risks and disclosure of personal
information.cognitive knowledge, indicating an individual's level of familiarity
with an issue (Nunes et al., 2011). Individuals gain knowledge from media,
education and personal experience. People who are heavily involved in a topic
tend to think that others are more affected by media messages because of their
limited knowledge (Xie and Johnson, 2015). For example, a person who
understands media risk will think that others understand through contact and
social exchange, this understanding is like the Domino effect, just one impact
factor, communication will spread around. A person perceives that the risk that
his or her private information or personal information being disclosed through
the media will negatively affect themselves, they tend to limit the disclosure of
their personal information. or using fake information to protect themselves from
the risk, whereas a person who disregards the disclosure of personal
information, ie, a low awareness of privacy risks, acts of sharing information on
social networks. Assembly was enhanced and unintentionally shared. Just as the
Privacy Analysis theory holds that individuals always reasonably weigh the
benefits and potential risks of data disclosure decisions, however, reasonable
assumptions often become erroneous. For example, they could be your current
feelings or moods. In addition, there are thinking styles such as perceived need
or belief in intuition. Framing a message, status bias, anchor effects, positive
25
bias or peer pressure (overwhelming impact and emotional commitment) can
lead to disclosure decisions. Unreasonable data. Therefore, the knowledge and
understanding of social media users about privacy risk is the factor that adjusts
the user's behavior to match the perception of the level of risk they consider to
have can happen to myself.
*Hypothesize:
H5:Users' perceptions of privacy risks on Facebook and Instagram are
positively influenced by user knowledge.
3.6. Effect of perception of privacy risk on personal information disclosure

behavior:
Through reference studies, we found that risk perception has an important
influence on the intention to disclose personal information, so we included this
factor in the study. This reference study has shown that privacy concerns are
negatively correlated with intention to disclose personal information, but this
effect varies in different contexts.
Privacy concerns have a negative correlation with self-disclosure. However, this
impact varies in different contexts; for example, Beuker established that privacy
concerns have less influence on the willingness to disclose personal information
in Dutch Facebook users. Contrarily, in a very recent study, Zhang and Fu
found that the volume, intimacy, and honesty of self-disclosure on social
networking sites have been adversely associated with privacy concerns.
Similarly, Kalmer and Schultheiss carried out a cross-cultural comparison of
Germans and Norwegian Facebook users on privacy and personal information
disclosure. They found that privacy concerns and attitudes have an indirect
effect on personal information sharing or self-disclosure.
In the vein of Youn (2009) and Dinev & Hart (2004), this article focuses on the
potential detrimental effects of knowledge sharing rather than the threats that
consumers of OSNs are aware of. As a result, perceived risks are described as
"the perceived negative consequences that an individual can face as a result of
disclosing personal information." According to Beldad et al. (2011, p. 222), the
threats associated with personal information exposure are many, and the risks
vary depending on the quantity and quality of information revealed. Individuals
are widely aware of the dangers of privacy invasions (e.g. Staksrud &
Livingstone, 2009) and the threats of unwanted access to data, even though the
risks that online information faces remain more unclear (e.g. Rezgui,
26
Bouguettaya & Eltoweissy, 2003). People are commonly aware that personal
information is often used for financial gain on the internet (e.g. Olivero & Lunt,
2004).
Inadequate data security (e.g., Youn, 2005) is another risk that causes fear.
Finally, OSN users are becoming more mindful that the details they freely share
can be misused by crooks, stalkers, bullies, or even one's own friends (e.g.
Staksrud & Livingstone, 2009; Saunders and Zucker, 1999). Because of the
possibility of being discovered online, those who reveal information online are
not aware of the real-world implications of their acts (Lee, Im & Taylor, 2008;
Youn, 2005). This may explain why people's trust in transparency decreases as
the sensitivity of the information sought rises (Castaeda & Montoro, 2007).
Users still do not understand the full risks of the details they disclose, according
to research (Dwyer, 2007; Govani & Pashley, 2005). According to Youn
(2005), "as teen privacy threats became more serious, they became less able to
provide personal information to a website."
*Hypothesize:
H6: The perception of privacy threats has a negative impact on personal
information disclosure.
4. CONCLUSION THEORY FOR GROUP TOPIC:

- We discovered that in relevant scientific publications, there was no particular
reference of which marketing theory had been applied, so we are rather cautious
in which theory to apply correctly via the study and investigation of marketing
theories as well as research articles linked to the group's issue. As a result, the
suitable theory is mostly chosen depending on whether the concept and
theoretical framework are relevant to the group's issue. In particular, our team
chose to use the Theory of Planned Behavior and Theory of Perceived Riskto
the modeling approach.
- As a consequence, we established two primary factors of the group topic based
on the risk perception of privacy theory: perceived danger of privacy and
intention to divulge personal information. Simultaneously, based on the theory
of expected behavior as well as models from relevant research publications,
choose 5 factors that have a significant impact on the topic's two key variables:
Trust, Perceived Benefits, Habit, Personalization, Knowledge.
27
-This broadens the scope of the topic's tentative model. The hypotheses
pertaining to the link between the variables are then built based on the
hypotheses as well as the two theories that we have chosen.
Figure 11. Research model
28
CHAPTER 3: RESEARCH DESIGN
1. RESEARCH PROCESS:
Theoretical basis
↓
Research models
↓
Qualitative research
↓
Complete Questionnaire
↓
Quantitative research
↓
Test Cronbach's Alpha & EFA
scale
↓
CFA analysis
↓
SEM analysis
↓
Analysis & discussion
↓
Conclude
29
2. QUALITATIVE RESEARCH:
Qualitative research is a form of exploratory research in which information is
collected in qualitative form through discussion and interpretation techniques
(Nguyen Dinh Tho and Nguyen Thi Mai Trang, 2007). The qualitative research
phase aims to evaluate:
- Assess the suitability of the scales after adjusting, adding observed variables to
the scale of the user's perceived value for the disclosure of personal information
through two social networking sites Facebook and Instagram.
- Check the word usage in each question of each observed variable to ensure
that all subjects in the surveyed age group understand the correct and clear
meaning.
- Preliminary examination of the correlation of the hypotheses given in the
research model.
Qualitative research results are the basis for building questionnaires for
quantitative research. The scales are inherited from previous studies, however,
different risk perception factors will have different characteristics.
In this study, qualitative research is represented by in-depth interviews with 8
people, including 1 person who is an expert on privacy issues. The remainder
were ordinary users of the social networks Facebook and Instagram,
interviewed using a set of questionnaires to uncover new factors influencing
information disclosure and perceived risk about information disclosure. privacy
as well as adjusting and supplementing the scales.
2.1. How to do it:

- Conduct in-depth interviews between group members and qualitative research
participants to collect relevant data, including 2 subjects:
Audiences are privacy experts: The choice of this audience is because this is a
person with expertise, qualifications, deep understanding of privacy issues, to
determine what factors affect users' sharing of personal information on the
website. social and what factors influence users' perception of privacy.
Audience is the focus group: Select about 7 people as friends to interview. The
common feature of these subjects is that they are students and students in
different schools, belong to the young population, use social networks regularly
to understand the perception and behavior of users about sharing information.
personal information and privacy risks.
30
-The group will interview 2 groups of subjects by online interviewing based on
a pre-established outline, the interview process takes place independently and in
groups. First, the group will interview a general question about the factors to be
surveyed with 2 groups of subjects. This question will contribute to the
perception, thinking and perception of each group of subjects that need to be
studied about the perception and intention to disclose personal information on
social networks. As for the risk factor, it will be the question of what risks do
you think users of social networks Facebook and Instagram will face when
disclosing personal information on social networks,... Then , will go specifically
into the observed variables of the factors to be investigated. Through the above
two steps, it will help the group to recognize the thoughts, perceptions, and
evaluations of the two research groups and the information they need to know
about the intention to disclose personal information as well as the perception of
risks about privacy. on social networks Facebook and instagram.
-Scale design:
The scale used to measure the factors in this study is built on the basis of TPR
risk perception theory, consumer behavior theory and planned behavior theory.
2.2. Results of qualitative research
2.2.1. Research model after qualitative research:

Experts and a focus group of interviews have suggested two further aspects that
influence perceptions of privacy risks and intention to disclose personal
information: privacy invasion experienceand privacy policy, based on
qualitative analysis. As a result, the team used the information to read more
research papers on these two variables, then built theories and added them to the
model to construct the most complete model possible.
A. Effect of privacy invasion experience on risk perceptions:

When people encounter a similar situation again, they normally equate their
previous experiences with future expectations. In general, social media sites are
held accountable for the security of their users' personal information. When
users discover that their personal information is being misused, they are likely
to believe that the platform is unable to meet its obligations. conducted a large-
scale investigation and discovered that, as opposed to those who have never had
their privacy violated, those who have had their privacy violated pay more
attention to information privacy and have a more negative attitude when asked
31
to provide personal information.Furthermore, several observational surveys
have shown that a privacy invasion experience would raise privacy concerns.
The attitude toward private information is influenced by prior experiences (Xu,
2007). When people who have previously been subjected to a breach of privacy
are confronted with a similar situation, their previous encounters are likely to
affect their potential perceptions (Zhao, Lu & Gupta, 2012). It has been
discovered that an online consumer's interpretation of a single online seller's
contract violation will affect the perception of contract violation by the entire
population of online sellers in the form of online marketplaces (Pavlou &
Gefen, 2005). Similarly, as people discover that their knowledge is being
misused on social media sites, they are more likely to believe that the platform
is not meeting its duties and may be more cautious in the future (Phelps, Nowak
& Ferrell, 2000). As a result, people who have been subjected to privacy
invasions are more concerned with privacy violations than those who have not.
*Hypothesize:
H7: Experiences of invasions of privacy have a major effect on risk
perceptions.
B. Effect of privacy policy on intention to disclose personal information:

One kind of privacy security is a privacy policy. It is described as a system that
informs users of their options for how the collected information is used; the
protections in place to protect the information from deletion, abuse, or
alteration; and how users can update or fix any incorrect information (Hui et al.,
2007). On this basis, Xu et al. (2011) discovered that users' perceived privacy
threats are reduced when privacy policies are perceived to be successful. Other
research (Chang et al., 2018; Culnan & Armstrong, 1999; Culnan & Bies, 2003;
Xu et al., 2008) reached similar conclusions. Furthermore, Culnan and
Armstrong (1999), Van Slyke, Shim, Johnson, and Jiang (2006) concluded that
providing a privacy policy to consumers through an entity limits users' self-
limitation in sharing personal information.
*Hypothesize:
H8: The perceived effectiveness of a privacy policy reduces intention to
disclose personal information on Facebook and Instagram.
32
Figure 12. Research model after qualitative research:
2.2.2. Scale adjustment for the topic:

Based on the completed research model, the team has built and adjusted the
scales corresponding to the following factors:
A. Scale of “trust”
The notation for trust is: TRUST. Seven observations are used to measure this
concept, denoted from TRUST1 to TRUST7. The observed variables of this
scale are measured by the seven-point interval scale and based on the scale of
Krasnova et al., 2010. The results of the trust scale include the following
observed variables:
I trust that Facebook and Instagram have sufficient expertise to handle my
Facebook and Instagram have good intentions about my personal information.
I trust that my Facebook and Instagram friends have the expertise not to
jeopardize my personal information.
I trust that my Facebook and Instagram friends do nothing with my personal
information that I will not consent to.
33
I trust that my Facebook and Instagram friends keep in mind my preferences
and desires regarding my personal information.
I trust that Facebook and Instagram protect my personal information from
companies and advertisers who want to misuse my information.
I believe Facebook and Instagram will check if people are playing by the rules.
B. Scale of “Perception of benefits”

Perceived benefit is denoted by PEB. Eight observations are used to measure
this concept, denoted PEB1 through PEB8. The observed variables of this scale
are measured using the octave scale and are based on the scale of Ellison et al.,
2006, p. 1151, and Krasnova et al., 2010. The results of the perceived benefit
scale include the following observed variables:
Facebook and Instagram are very useful for exchanging personal information
with your friends.
By sharing personal information on Facebook and Instagram, I get to know
people better.
Facebook and Instagram are great for me to keep track of what other people
share about them.
Sharing personal information on Facebook and Instagram is fun.
On Facebook and Instagram, I have more courage in sharing personal
information than in other situations.
By sharing personal information on Facebook and Instagram, I become more
popular with my Facebook friends.
I share personal information via Facebook and Instagram because it is better
than the alternatives.
By sharing personal information on Facebook and Instagram, I can make a good
impression on my Facebook friends.
C. Scale of “Habits”
Habit is denoted by HA. Four observations were used to measure this concept,
denoted HA1 to HA4. The observed variables of this scale are measured by the
octave scale and are based on the scale of Verplanken & Orbell, 2003, p. 1329.
The results of the habit scale include the following observed variables:
Sharing personal information is a habit for me.
I sometimes share personal information without thinking about it.
34
Sometimes I share personal information because it's hard for me not to share it
with others
If I see something interesting, the first thing that comes to mind is to share it on
Facebook.
D. Scale of “Personalization”
Personalize the symbol as PE. Three observations are used to measure this
concept, denoted PE1 through PE3. The observed variables of this scale are
measured by the seven-point interval scale and are based on the scale of Xu,
Luo, Carroll & Rosson, 2011. The results of the personalized scale include the
following observed variables:
Facebook and Instagram may provide me with personalized promotions and
deals.
Facebook and Instagram provide me with a push message respectively based on
my interests or personal preferences.
Facebook and Instagram provide me with ads that I like.
E. Scale of “Knowledge”
Notational knowledge is KNL. Six observations were used to measure this
concept, denoted KNL1 through KLN6. The observed variables of this scale are
measured using the seven-point interval scale and are based on the scale of Xu,
Luo, Carroll & Rosson, 2011. The results of the knowledge scale include the
How knowledgeable are you about virus attacks?
How knowledgeable are you about hacker attacks?
How knowledgeable are you about identity theft?
How knowledgeable are you about credit card theft?
How well do you understand privacy invasions?
How knowledgeable are you about online insults?
F. Scale “Privacy risk perception”

Privacy risk perception denotes PPR. 11 observations were used to measure this
concept, denoted PPR1 through PPR11. The observed variables of this scale are
measured by the octave scale and are based on the scale of Based on Diev
35
&Hart, 2004. The results of the privacy risk perception scale include the
Social networking sites Facebook and Instagram are a danger to the security of
my personal information
I believe that Facebook and Instagram sell my personal information to others.
I am afraid that Facebook and Instagram secretly use my personal information
for purposes with which I do not agree.
Friends of friends or companies on Facebook and Instagram are a danger to my
Hackers are a danger to my personal information on Facebook and Instagram.
I'm afraid my Facebook and Instagram friends will get the wrong impression of
me because of the personal information I've shared on Facebook
My Facebook and Instagram friends are a danger to the safety of my personal
information.
I'm afraid my Facebook and Instagram friends might use my personal
information to bully me.
I don't care about the risks of sharing personal information.
I’m afraid that I have unintentionally shared personal information because I
made a mistake.
There are dangers in sharing personal information without my knowledge.
G. Scale of “Privacy Intrusive Experience”

Practical experience denoted by PREX. Four observations were used to measure
this concept, denoted from PREX1 to PREX4. The observed variables of this
scale are measured by the octave scale and are based on the scale of Xu, Diev,
Smith & Hart, 2011a and Li, K.; Wang, X.; Li, K.; Che, J; 2016. The results of
the privacy invasion experience scale include the following observed variables:
How many times do you feel your privacy has been violated?
How many times have you heard or read in the past year about consumer
privacy abuses?
In the past year, how many times have you heard about the indiscriminate use of
personal information on the Internet?
How often are you the victim of an invasion of privacy?
36
H. Scale of “Privacy Policy”
The privacy policy is denoted by POLICY. Three observations were used to
measure this concept, denoted from POLICY1 to POLICY3. The observed
variables of this scale are measured using the octave scale and are based on the
scale of Chang et al. (2018), Xu et al. (2011). The results of the security policy
scale include the following observed variables:
With the privacy statements, on Facebook and Instagram, I believe that my
personal information will be kept private and confidential.
I believe the privacy statements on Facebook and Instagram reflect their
commitment to protecting my personal information.
I believe the privacy statements on Facebook and Instagram are an effective
way to demonstrate their commitment to privacy.
I. Scale of “Intent to disclose personal information”

The intention to disclose personal information is denoted by DPI. Fifteen
observations are used to measure this concept, denoted from DPI1 to DPI16.
The observed variables of this scale are measured by the octave scale and are
based on the scale of Ruud H.G. Koehorst, 2013. The results of the scale of
intention to disclose personal information include the following observed
variables:
I filled the city I live in.
I filled in the address where I work.
I filled in the blanks with my family members.
I will update if I have a relationship.
I gave my phone number.
I often share my opinions on Facebook and Instagram.
I often share on Facebook and Instagram where I am at the time.
The qualitative research results show that the research model "Privacy risk
perception and disclosure of personal information by users of social networks
Facebook and Instagram" has two emerging factors that the interviewees
considered to affect the privacy risk perception and disclosure of personal
information by Facebook and Instagram users. It is an invasion of privacy and
security policy. The observed variables used for the concept of measuring users'
perceptions and intentions about privacy and personal information disclosure
are measured using a 7-point Likert scale. After conducting interviews,
37
adjusting and supplementing from the preliminary scale, the results have 61
observed variables used to measure for nine concepts in the research model.
Table 1: Results of qualitative research and scale
No Measurement items Items
TRUST
1 I trust that Facebook and Instagram have the TRUST1
expertise to handle my personal information.
2 Facebook and Instagram have good intentions for TRUST2
my personal information.
3 I trust that my Facebook and Instagram-friends TRUST3
have the expertise to not jeopardize my personal
information.
don’t do anything with my personal information I
would not approve of
keep my preferences and desires about my personal
information in mind.
6 I trust that Facebook and Instagram protect my TRUST6
personal information against companies and
advertisers that want to abuse my information.
7 I trust that Facebook and Instagram check if TRUST7
everybody is playing by the rules.
PERCEIVED BENEFITS
8 Facebook and Instagram are useful to exchange PEB1
personal information with your friends.
9 Thanks to sharing personal information on PEB2
38
Facebook and Instagram, I get to know people
better.
10 Facebook and Instagram are useful for me to PEB3
monitor what others share about themselves.
11 Sharing personal information on Facebook and PEB4
Instagram is fun.
12 On Facebook and Instagram I have more courage PEB5
in sharing personal information compared to other
situations.
13 By sharing personal information on Facebook and PEB6
Instagram, I get more popular with my Facebook-
friends.
14 I share personal information via Facebook and PEB7
Instagram because it’s better than the alternatives.
15 By sharing personal information on Facebook and PEB8
Instagram, I can make a good impression on my
Facebook-friends.
HABITS
16 Sharing personal information is a habit to me. HA1
17 I sometimes share personal information without HA2
thinking about it.
18 I sometimes share personal information because it HA3
is hard for me not to share it with others
19 If I see something interesting, the first thing that HA4
comes into mind is to share it on Facebook.
PERSONALIZATION
20 Facebook and Instagram can provide me with PE1
39
personalized advertising and transactions.
21 Facebook and Instagram provide me with a PE2
corresponding push message based on my
preferences or personal interests.
22 Facebook and Instagram provide me with ads I PE3
like.
KNOWLEDGE
23 Perceived understandings about virus attacks. KNL1
24 Perceived understanding about hacker attacks. KNL2
25 Perceived understandings about identification theft. KNL3
26 Perceived understandings about credit card theft. KNL4
27 Perceived understandings about privacy invasion. KNL5
28 Perceived understanding about online insults. KNL6
PERCEPTION OF PRIVACY RISKS
29 Facebook and Instagram as a company is a danger PPR1
for the safety of my personal information.
30 I’m afraid that Facebook and Instagram sell my PPR2
personal information to others.
31 I’m afraid that Facebook and Instagram secretly PPR3
use my personal information for purposes I don’t
agree with.
32 Friends-of-friends or companies on Facebook and PPR4
Instagram are a danger to my personal information.
33 Hackers are a danger for my personal information PPR5
on Facebook and Instagram.
34 I’m afraid my Facebook and Instagram-friends get PPR6
a wrong impression of me because of the personal
information I’ve shared on Facebook.
35 My Facebook and Instagram-friends are a danger PPR7
40
to the safety of my personal information.
36 I’m afraid my personal information can be used by PPR8
my Facebook and Instagram friends to bully me
with.
37 I don’t care about the risks of sharing personal PPR9
information.
38 I’m afraid that I unintentionally shared personal PPR10
information because I made a mistake.
39 There are dangers to sharing personal information PPR11
that I’m not aware of
INTENTION TO DISCLOSE PERSONAL
INFORMATION
40 I filled in what town I live in. DPI1
41 I filled in where I work. DPI2
42 I filled in who my family members are. DPI3
43 I am filled in if I have a relationship. DPI4
44 I filled in my phone number. DPI5
45 I often share my opinion on Facebook and DPI6
Instagram.
46 I often share on Facebook and Instagram where I DPI7
am at that moment.
PRIVACY INVASION EXPERIENCE
47 How many times have you felt that your privacy PREX1
has been violated?
48 In the past year, how many times have you heard PREX2
about the indiscriminate use of private information
on the Internet?
49 How much have you heard or read during the last PREX3
41
year about the use and potential misuse of
information privacy about consumers?
50 How often have you personally been a victim of PREX4
what you felt was an invasion of privacy?
PRIVACY POLICY
51 I feel confident that privacy statements on POLICY1
Facebook and Instagram reflect their commitments
to protect my personal information.
52 With privacy statements, on Facebook and POLICY2
Instagram I believe that my personal information
will be kept private and confidential.
53 I believe that privacy statements on Facebook and POLICY3
Instagram are an effective way to demonstrate their
commitments to privacy.
2.3. Questionnaire design

After completing the adjustment and building the appropriate scales for
surveying the factors affecting privacy perception and intention to disclose
personal information on social networks Facebook and Instagram, the team
conducted Design a questionnaire for data collection. The formal questionnaire
used in quantitative research consists of two parts:
PART 1: QUESTIONS ABOUT THE DEMOGRAPHY
Record the information the user is male or female, how old is, what is their
occupation, whether they use Facebook and Instagram, how often they use
them, how long they use them, and how many friends they have. This is the
information for descriptive analysis.
PART 2: QUESTIONS FOR VARIABLE RESEARCH.
Record the degree of agreement observed variables (expressed by statements)
measure the concepts in the model. This is also the main part of the
questionnaire. In this section, we give questions measured on a scale from 1-7,
corresponding to the answer items from Completely disagree - Totally agree,
42
Completely do not understand - Completely understand, Not yet ever - Many
times.
3. QUANTITATIVE RESEARCH:
3.1. Sampling method

In this study, the sampling method chosen is convenience sampling, the data is
collected through the form of submitting an online survey form.
3.2. Sample size

Method of calculating the number of samples by age clusters:
Regarding the age to be surveyed, we decided to focus mainly on the age group
15 to 29 and divided into the following ranges:
From 15 to 19 years old.
Because the method of calculating the number of samples is accurate and the
number of people of each age is divided according to the recent total report on
the population of Vietnam in 2019.
In 2019, Vietnam's population is 96,462,108 people.
Because the gaps need to be surveyed, groups focus on only 3 main groups
stretching from 15 to 29 years old.
We have the formula for calculating the number of samples:
n: Number of sample members to be determined for the investigation study.

N: The total number of samples.
e: Standard deviation and e = 0.05
So we have:
N: Vietnam's total population will be between 15 and 29 years old in 2019. And
N= 22,361,416 people.
43
Table 2: Calculate the number of survey samples
Age Vietnam's population in Number of sample
2019 by age group members on each group
From 15 to 19 years old. 6,474,991 116
From 20 to 24 years old. 7,145,151 128
From 25 to 29 years old. 8,741,274 156
Total 22,361,416 400
In this study, the minimum number of samples needed is 400. By submitting the
online form, the team obtained 428 answer sheets. After performing the test, out
of 428 responses, 400 were selected to be included in the study.
3.3. Processing and analyzing data

The collected data will be encrypted and analyzed as follows:
Verification and evaluation of the scale
To evaluate the scale of concepts in the research, it is necessary to check the
reliability and validity of the scale. Based on the Cronbach's Alpha reliability
coefficients, the item-to-total correlation coefficient helps to exclude the
observed variables that do not contribute to the description of the concept to be
measured, the coefficient Cronbach's alpha if Item Deleted to help evaluate and
remove observed variables to improve the reliability coefficient of
Cronbach'sAlpha for the concept to be measured and the exploratory factor
analysis (EFA) method to check the validity of the scale of the concepts.
research concept.
Cronbach's Alpha Analysis

Using Cronbach's Alpha reliability coefficient method before analyzing EFA
factors to eliminate inappropriate variables because these variables can create
44
dummy factors (Nguyen Dinh Tho & Nguyen Thi Mai Trang, 2007). Criteria
used when assessing the reliability of the scale: Type of observed variables with
small variable-total correlation coefficient (less than 0.3). According to Nunnall
& Burnstein (1994) that variables with variable-sum correlation coefficient less
than 0.3 are considered garbage variables and will be excluded from the model;
criteria for choosing the scale when the alpha reliability is greater than 0.6 (the
larger the alpha, the higher the internal consistency reliability) (Nunnally &
Burnstein 1994; Nguyen Dinh Tho and Nguyen Thi Mai Trang, 2007).
Exploratory factor analysis (EFA)

-After removing the variables that do not guarantee reliability, the EFA factor
analysis method is used to determine the convergent validity, discriminant
validity, and at the same time collect compact the estimated parameters for each
group of variables.
-In order for the scale to reach convergent value, the single correlation
coefficient between variables and factors (factor loading) must be greater than
or equal to 0.4 in a factor (Jun & Ctg 2002). To achieve discriminant validity,
the difference between the factors must be greater than or equal to 0.3 (Jabnoun
& Ctg 2003). The number of factors is determined based on the Eigenvalue
index - representing the variation explained by each factor. According to Kaiser
criteria, factors with Eigenvalue less than 1 will be excluded from the model
(Garson 2003). Variance explained criteria: the total variance extracted must be
greater than 50%. Considering the KMO value: 0.5 < KMO< 1, the factor
analysis is suitable for the data; whereas KMO<0.5, factor analysis is likely not
suitable for the data (Hoang Trong & Chu Nguyen Mong Ngoc, 2005). Using
Principal axis factoring method with Promax rotation (kappa = 4) and stopping
point when extracting factors with Eigenvalues greater than 1 for observed
variables measuring 9 component concepts affecting perception of the user.
According to this method the hidden dimensions (components) are detected in
the original data (structure detection). Choosing the Promax rotation will more
accurately reflect the underlying data structure. Using the method of factor
extraction Principal components with Varimax rotation and stopping point when
extracting factors with Eigenvalues greater than 1 with observed variables
measuring the concept of usage trend.
Confirmatory factor analysis (CFA)
45
After running the general EFA for the data table, it will move to CFA. In CFA
confirmatory factor analysis, the evaluation of model fit is extremely important.
The scales and latent variables form a measurement model of the concepts used
in the research. With a collected data set, we need to test whether this
measurement model with input data meets the requirements. Do the scales meet
the requirements of a good scale? Any observed variable that does not
contribute to the model does not. For this we need to use confirmatory factor
analysis (CFA).
According to Hair et al. (2010), Multivariate Data Analysis, 7th edition of the
indicators considered to evaluate Model Fit while running CFA include many
discriminant indicators such as: CMIN/df index less than 2 is good; CFI index
greater than 0.8 is acceptable; GFI index greater than or equal to 0.9 is good (In
some topics, due to the limitation of sample size, it is difficult for the GFI value
to reach 0.9 because this index depends a lot on the number of scales, the
number of observed variables and the sample size. Therefore, if the GFI value is
below 0.9 but from 0.8 or higher, it is still accepted according to 2 studies by
Baumgartner and Homburg (1995) and Doll, Xia, and Torkzadeh (1994));
RMSEA index less than 0.08 is good.
Structural Equation Modeling (SEM)

-SEM is a second generation statistical analysis technique developed to analyze
multidimensional relationships between many variables in a model (Haenlein &
Kaplan et al. , 2004). Multiple relationships between variables can be
represented in a variety of simple and multiple regression equations. The
structural linear modeling technique uses the combination of quantitative data
and correlated (cause-effect) assumptions into the model. With SEM,
researchers can visually examine the relationships that exist between variables
of interest to prioritize resources to better serve customers. The fact that latent
variables that are difficult to measure can be used in SEM makes it ideal for
solving business research problems. SEM is a more powerful statistical
technique to address the following claims: Analyze multiple regression models
simultaneously; Regression analysis with multicollinearity problem; Path
analysis analysis with many dependent variables; Model multidimensional
relationships between variables in a model.
46
CHAPTER 4: QUANTITATIVE RESEARCH
RESULTS
1. SAMPLE DESCRIPTION:
By submitting the online form, the team obtained 428 answer sheets. After
performing the test, out of 428 answer sheets, 400 votes were selected to be
47
included in the study and run the data, the sample classification structure
according to the following criteria
1.1. Sample structure by sex

Out of 400 valid responses, 147 respondents were male (36.8%), 252
respondents were female (63%), 1 other vote accounted for (0.2%).
Table 3: User's gender
Frequency Percent Valid Cumulative

Percent Percent
Valid Khac 1 .2 .2 .2
Nam 147 36.8 36.8 37.0
Nu 252 63.0 63.0 100.0
Total 400 100.0 100.0
1.2. Sample structure by age group

Among 400 answers, there are 116 people from 15-19 years old, accounting for
29%. There are 128 people from 20-24 years old, accounting for 32%, and there
are 156 people from 25-29 years old, accounting for 39%.
Table 4: User's age

Percent Percent
Valid Tu 15-19 116 29.0 29.0 29.0

tuoi
Tu 20-24 128 32.0 32.0 61.0

tuoi
48
Tu 25-29 156 39.0 39.0 100.0
tuoi
Total 400 100.0 100.0
1.3. Sample structure by industry group:

Out of 400 responses received, there are:
11 owners and business leaders (2.8%).
Civil servants and public employees there are 33 people (8.3%).
Skilled workers have 8 people (2%).
Self-sales service has 9 people (2.3%).
Small businesses have 17 people (4.3%).
Students have 231 people (57.8%).
10 people are not working (2.5%).
Normal labor has 18 people (4.5%).
Employees have 63 people (15.8%).
Table 5: User's occupation

Percent Percent
Valid Chu so huu, lanh 11 2.8 2.8 2.8

dao doanh nghiep
Cong chuc,vien 33 8.3 8.3 11.0

chuc
Cong nhan lanh 8 2.0 2.0 13.0

nghe
49
Dich vu tu ban 9 2.3 2.3 15.3
hang
doanh nghiep nho 17 4.3 4.3 19.5
Hoc sinh, sinh vien 231 57.8 57.8 77.3
Khong lam viec 10 2.5 2.5 79.8

(Noi tro, that
nghiep,...)
Lao dong binh 18 4.5 4.5 84.3

thuong
Nhan vien 63 15.8 15.8 100.0
Total 400 100.0 100.0
1.4. Sample structure according to usage habits:

In more than 400 group samples collected, 100% of users use both social
networks: Facebook and Instagram. Checking if the survey sample uses both
Facebook and Instagram at the same time, makes the results more accurate and
transparent.
Table 6: Using Facebook and Instagram

Percent Percent
Valid Yes 400 100.0 100.0 100.0
1.5. Sample structure according to frequency of use:

Out of 400 selected answer sheets, there are people uses:
Less than 1 hour per day: 48 people (12%).
More than 5 hours per day: 95 people (23.75%).
About 2-3 hours per day: 130 people (32.5%).
50
About 3-5 hours per day: 127 people (31.75%).
Table 7: Frequency of Use Facebook and Instagram

Percent Percent
Valid It hon 1 gio 48 12.0 12.0 12.0
Nhieu hon 5 95 23.8 23.8 35.8

gio
Tu 2 - 3 gio 130 32.5 32.5 68.3
Tu 3 - 5 gio 127 31.7 31.7 100.0
Total 400 100.0 100.0
1.6. Sample structure by number of friends:

More than 1000 friends: 106 people (26.5%)
Less than 100 friends: 22 people (5.5%)
About 100-500 friends: 118 people (29.5%)
About 500-1000 friends: 154 people (38.5%)
Table 8: a user's number of Facebook and Instagram friends

Percent Percent
Valid Hon 1000 106 26.5 26.5 26.5
It hon 100 22 5.5 5.5 32.0
Khoang 100- 118 29.5 29.5 61.5

500
Khoang 500- 154 38.5 38.5 100.0
51
1000
Total 400 100.0 100.0
1.7. Sample structure by time of use:

Over 5 years: 185 people (46.3%)
Less than 1 year: 23 people (5.8%)
From 2-3 years: 66 people (16.5%)
From 3-5 years: 126 people (31.5%)
Table 9: Used time Facebook and Instagram

Percent Percent
Valid Hon 5 185 46.3 46.3 46.3

nam
It hon 1 23 5.8 5.8 52.0

nam
Tu 2-3 66 16.5 16.5 68.5

nam
Tu 3-5 126 31.5 31.5 100.0

nam
Total 400 100.0 100.0
52
2. CHECK THE SCALE:
The latent variables in the model before conducting exploratory factor analysis
are tested for the reliability of the scale. To test the reliability of the scale, the
research team used Cronbach's Alpha coefficient. The minimum Cronbach's
Alpha selection criterion is 0.6 according to Hair et al., 1998, but the research
team has taken the minimum standard for the Cronbach's Alpha study to be 0.7
and the correlation coefficient of the sum of the observed variables is minimum.
is 0.3 according to Nunally and Burstein, 1994. The results of the scale test of
the research variables with the lowest Cronbach's Alpha is 0.707 of the variable
"Personalization" and the highest Cronbach's Al phase is 0.87 of the variable
"Risk perception" privacy risk”, the specific results are as follows:
2.1. Testing the scale of “Belief”

Factor "Trust" in the model is measured by 7 observed variables, the test results
using SPSS 20.0 show that Cronbach's Alpha coefficient is 0.858 > 0.7. The
correlation coefficient of the total variable of 7 observed variables from
TRUST1 to TRUST7 is all 0.3, so no observed variables are excluded.
Table 10: Reliability Statistics and Item-Total Statistics

Cronbach’s Alpha N of Items
.858 7
Scale Mean if Scale Variance if Corrected Item- Cronbach's
Item Deleted Item Deleted Total Correlation Alpha if Item
Deleted
TRUST1 26.99 46.757 .621 .838
TRUST2 27.28 44.976 .653 .833
TRUST3 27.16 45.079 .654 .833
TRUST4 27.21 43.976 .620 .839
TRUST5 27.37 45.892 .558 .847
TRUST6 27.47 44.120 .637 .836
TRUST7 27.15 45.707 .627 .837
53
2.2. Testing the scale of “Perception of benefits”
The factor "perceived benefit" in the model is measured by 8 observed
variables, the test results show that Cronbach's Alpha is 0.860 greater than 0.7
and the correlation coefficient between the sum of the 8 observed variables is
satisfactory. is greater than 0.3 so no observed variables are excluded.

.860 8
Deleted
PEB1 35.11 48.186 .481 .856
PEB2 35.36 46.000 .632 .841
PEB3 35.20 45.928 .601 .844
PEB4 35.44 44.342 .625 .841
PEB5 35.81 43.270 .669 .835
PEB6 35.58 44.039 .628 .840
PEB7 35.84 44.337 .596 .844
PEB8 35.59 45.075 .616 .842
2.3. Testing the scale of “Habit”

The factor "Habit" in the research model is measured by 4 observed variables
from HA1 to HA3, the test results show that Cronbach's Alpha is 0.853, greater
than 0.7 according to the research group's standards and has a correlation
54
coefficient between variables. sum is greater than 0.3, all observed variables
have a great influence on the reliability of the scale and no observed variables
are excluded.

.853 4
Deleted
HA1 12.23 19.314 .757 .787
HA2 12.58 18.424 .730 .797
HA3 12.24 19.490 .690 .815
HA4 11.83 21.281 .604 .849
2.4. Testing the scale of the “Personalization”

The PE factor is measured by 3 observed variables PE1 to PE3, the results of
testing the Reliability show that Cronbach's alpha coefficient is 0.707 > 0.7.
And the correlation coefficient of the total variable is greater than 0.3. There are
no unsatisfactory variables, the specific results are as follows

.707 3
Deleted
PE1 9.81 6.047 .541 .599
55
PE2 9.64 6.311 .519 .628
PE3 9.99 5.090 .528 .624
2.5. Testing the scale of “Knowledge”

The factor "Knowledge" is measured by 6 observed variables from KNL1 to
KNL6. The results of the reliability test show that Cronbach's alpha coefficient
is 0.86 > 0.7. And the correlation coefficient of the total variable is greater than
0.3. Details are shown in the table below.

.860 6
Deleted
KNL1 23.99 32.150 .646 .838
KNL2 23.96 31.484 .731 .821
KNL3 23.86 31.198 .770 .814
KNL4 23.78 32.470 .665 .834
KNL5 23.26 35.628 .594 .847
KNL6 23.01 37.551 .508 .860
2.6. Testing the scale of “Privacy Invasions Experience”

The factor "Privacy Invasions Experience" is measured by 4 observed variables
from PREX1 to PREX4, the reliability test results show that the Cronbach's
Alpha coefficient is equal to 0.488 < 0.6 the minimum value. That shows that
the scale is not reliable enough to measure. In which, we see that 2 observed
variables PREX1 and PREX4 have total correlation coefficients of 0.183 and
0.116 < 0.3, respectively, so 2 variables PREX1 and PREX4 are considered as a
bad variable and removed from the PREX scale.
At the same time, the removal of 2 variables helps to increase the coefficient of
cronbach's alpha to 0.786 > 0.7. Therefore, to ensure reliability, PREX will be
56
measured by 2 observed variables PREX2 and PREX3. Details are given in the
table below:

.786 2
Deleted
PREX2 4.96 2.061 .648 .
PREX3 4.80 2.062 .648 .
2.7. Testing the scale of "Privacy Policy"

The factor “Privacy Policy” is measured by 3 observed variables from
POLICY1 to POLICY3, the reliability test results show that Cronbach's alpha
coefficient is 0.826 > 0.7. And the correlation coefficient of all variables is
greater than 0.3, therefore, no observed variables are excluded and they have
high confidence. Details are shown in the table below.

.827 3
Deleted
POLICY1 9.05 6.659 .695 .750
POLICY2 8.88 7.001 .680 .765
POLICY3 8.76 6.733 .677 .768
2.8. Testing the intermediate variable scale “Privacy risk perception”

The intermediate variable PPR is measured by 11 observed variables from
PPR1 to PPR11, the reliability test results show that Cronbach's alpha is 0.87 >
0.7. And the correlation coefficient of the total variable is greater than 0.3. This
proves that the observed variables have high reliability.
57
.870 11
Deleted
PPR1 45.40 106.797 .597 .858
PPR2 45.41 103.220 .623 .855
PPR3 45.24 102.799 .670 .852
PPR4 45.77 103.976 .635 .855
PPR5 44.93 105.532 .596 .857
PPR6 45.39 104.153 .631 .855
PPR7 45.65 103.652 .656 .853
PPR8 45.72 102.660 .634 .854
PPR9 46.62 109.811 .323 .879
PPR10 45.59 106.038 .494 .865
PPR11 454.96 107.519 .502 .864
2.9. Testing the dependent variable scale “Intent to disclose personal

information”
The DPI dependent variable is measured by 15 observed variables from DPI1 to
DPI15, the reliability test results show that Cronbach's alpha is 0.814 > 0.7.
However, including the observed variable DPI15, the total correlation
58
coefficient is 0.272<0.3. Removing the DPI15 observational variable and
running the factor discovery analysis again gave Cronbach's Alpha results of
0.812 and all the remaining observed variables had a total correlation coefficient
greater than 0.3. This proves that the observed variable DPI15 has low
confidence and it needs to be removed from the DPI variable.9 Testing the
dependent variable scale “Intent to disclose personal information”. The DPI
dependent variable is measured by 15 observed variables from DPI1 to DPI15,
the reliability test results show that Cronbach's alpha is 0.814 > 0.7. However,
including the observed variable DPI15, the total correlation coefficient is
0.272<0.3. Removing the DPI15 observational variable and running the factor
discovery analysis again gave Cronbach's Alpha results of 0.812 and all the
remaining observed variables had a total correlation coefficient greater than 0.3.
This proves that the observed variable DPI15 has low confidence and it needs to
be removed from the DPI variable.

.758 7
Deleted
DPI1 24.93 38.304 .550 .712
DPI2 24.76 39.766 .451 .735
DPI3 25.55 36.594 .607 .698
DPI4 24.70 37.088 .521 .719
DPI5 24.85 43.777 .394 .745
DPI6 24.44 43.234 .374 .749
DPI7 24.70 41.558 .431 .738
Table 19: The final summary table of the scale test:

Factor Cronbach’s Alpha N
TRUST 0.858 7
PEB 0.860 8
HA 0.853 4
59
PE 0.707 3
KNL 0.860 6
PREX 0.786 2
POLICY 0.827 3
PPR 0.870 11
DPI 0.758 7
3. EXPLORATORY FACTOR ANALYSIS (EFA):

After testing the scale with Cronbach's Alpha reliability coefficient, the scales
are next evaluated by factor analysis EFA. Principal components extraction
method with Varimax rotation is applied. The extracted factors are the smallest
according to Hoang Trong and Chu Nguyen Mong Ngoc, 2008. For this study,
factor discovery analysis was performed simultaneously with the variables.
Independent, dependent, and intermediate variables were analyzed separately.
Factor loading factor analysis standard is greater than 0.5, the eigenvalue is
greater than or equal to 1, minimum extracted variance is 50%, KMO
coefficient is greater than 0.5, Bartlett test has p-value (sig) less than 0.05,
Factor loading value suitable for sample size 400 is 0.3, specific analysis results
are as follows.
3.1.Exploratory factor analysis with independent variables

The data was conducted exploratory factor analysis by SPSS 20.0 for
independent variables, the observed variables were removed after testing the
scale (PREX1, PREX4), Exploratory Factor Analysis for the second time. as
follows: KMO coefficient is 0.884 and Bartlett's test has p-value equal to .000 <
0.05, extracted variance is 62.913% greater than 50%, eigenvalue value is 1,214
greater than 1. But the first-factor rotation matrix found that there are two bad
variables TRUST5 and KNL6, these two observed variables, have a Factor
loading effect of less than 0.3, so they have been removed.
60
Table 20: 1st- Rotated Component Matrix
Component
1 2 3 4 5 6 7
PEB5 .705
PEB7 .675
PEB4 .670
PEB2 .657
PEB3 .651
PEB6 .651
PEB8 .602
PEB1 .593
TRUST3 .745
TRUST1 .729
TRUST4 .711
TRUST2 .696
TRUST7 .695
TRUST6 .659 .310
TRUST5 .555 .388
KNL2 .861
KNL3 .853
KNL1 .804
KNL4 .734
KNL5 .643
KNL6 .519 .370
HA2 .811
HA1 .808
HA3 .740
HA4 .646
61
POLICY
.865
1
POLICY
.856
2
POLICY
.855
3
PE2 .758
PE1 .696
PE3 .684
PREX3 .857
PREX2 .852
The second-factor exploratory analysis is obtained after removing the bad

variable KNL6, TRUST5 as follows:
KMO coefficient is 0.876 > 0.5, exploratory factor analysis is consistent with
the model, and Bartlett's test has a p-value equal to .000 < 0.05, extracted
variance is 64.054%, greater than 50%, so the observed variable formed 7
factors that explained 64.162% of the variation of the total variable and had an
eigenvalue of 1.195 greater than 1, no negative variables appeared, specifically
see in the matrix of the 2nd-factor rotation.
2nd-factor rotation matrix:
Table 21: 2nd - Rotated Component Matrix

Component
1 2 3 4 5 6 7
PEB5 .706
PEB7 .675
PEB4 .675
PEB2 .663
PEB6 .658
PEB3 .657
PEB8 .609
PEB1 .591
TRUST3 .747
TRUST1 .733
TRUST4 .716
TRUST2 .695
TRUST7 .305 .694
TRUST6 .662 .324
KNL2 .871
KNL3 .859
KNL1 .812
KNL4 .735
KNL5 .621
62
HA2 .817
HA1 .813
HA3 .742
HA4 .655
POLICY1 .865
POLICY2 .856
POLICY3 .855
PE2 .766
PE1 .708
PE3 .692
PREX2 .872
PREX3 .869
3.2. Exploratory factor analysis of DPI ( Dependent variable) and ( PPR

Intermediate variable)
The results of exploratory factor analysis from research data using SPSS version
20.0 for the intermediate variable PPR and the dependent variable DPI, for the
first run, the results are satisfactory: KMO coefficient is 0.873 and Bartlett's test
has a p-value of .000 < 0.05, the extracted variance is 50.508% greater than
50%, the eigenvalue of 1,199 is greater than 1. However, with the first run
coefficient rotation matrix detecting the bad variable PPR9, this observation
variable is uploaded into 2 groups of factors. Therefore, PPR9 will be excluded
from the observed variables.
Table 22: 1st-rotation matrix

Component
1 2
PPR3 .765
PPR5 .731 -.326
PPR7 .714
PPR2 .710
PPR6 .703
PPR4 .700
PPR8 .695
PPR1 .670
PPR11 .652 -.415
PPR10 .584
DPI3 .759
DPI1 .706
DPI4 .697
DPI2 .615
63
DPI5 .587
DPI7 .565
DPI6 .502
PPR9 .323 .720
However, after removing the observed variable PPR9, the results are not
satisfactory because the extracted variances are all less than 50%. Therefore, the
group in turn selects the observed variables with lower Factor loading than other
variables and removes them to get better results. And after removing 3 bad
variables in the order of DPI6, DPI7, PPR10, there are results: KMO coefficient
is 0.864 and Bartlett test has p value of 0.000 < 0.05, extracted variance is
50.148% larger 50%, the eigenvalue is 2,413 which is greater than 1.
Table 23: 2nd-rotation matrix
Component
1 2
PPR3 .778
PPR7 .731
PPR2 .729
PPR4 .728
PPR5 .713
PPR6 .707
PPR8 .704
PPR1 .680
PPR11 .614
DPI3 .779
DPI1 .736
DPI4 .706
DPI2 .626
DPI5 .623
Table 24: Summary for results KMO:

Factor KMO Sig Cumulative
64
%
Independent variables 0.878 0.000 64.054
Intermediate and independent 0.864 0.000 50.148
variables
3.3. Exploratory factor analysis before running CFA and SEM:

For this study, factor discovery analysis was performed simultaneously with the
variables. Data extraction method: Principal Axis Factoring and the rotation is
Promax applied. Factor loading factor analysis standard is greater than 0.5, the
eigenvalue is greater than or equal to 1, minimum extracted variance is 50%,
KMO coefficient is greater than 0.5, Bartlett test has p-value (sig) less than
0.05, Factor loading value suitable for sample size 400 is 0.3, specific analysis
results are as follows.
After performing EFA for all factors, there are bad observed variables that need
to be eliminated. Removing each variable in turn will help us improve the
results as well as control the level of the problem. And after removing 3 bad
variables: KNL5, HA4, PEB1, the results are as follows: KMO coefficient is
0.849 and Bartlett test has p value of 0.000 < 0.05, extracted variance is
54.526% larger 50%, the eigenvalue is 1.011 which is greater than 1.
Table 25: Pattern Matrix

Factor
1 2 3 4 5 6 7 8 9 10
PPR7 .792 -.392
PPR3 .695
PPR6 .692
PPR8 .689
PPR4 .676
PPR2 .661 .359
PPR5 .661
PPR1 .625
PPR11 .547
PEB5 .756
PEB6 .701
PEB7 .662
65
PEB2 .636
PEB4 .632
PEB3 .600
PEB8 .551
TRUST3 .798
TRUST4 .762
TRUST1 .662
TRUST7 .612
TRUST2 .603
TRUST6 .559
KNL2 .884
KNL3 .834
KNL1 .766
KNL4 .625
HA1 .799
HA2 .775
HA3 .654
DPI3 .725
DPI4 .639
DPI1 .620
DPI5 .527
DPI2 .515
POLICY
.793
3
POLICY
.793
1
POLICY
.781
2
PE2 .664
PE1 .654
PE3 .554
PREX3 .830
PREX2 .758
Because the factor loading factor analysis standard is greater than 0.5. So
column number 10 will be removed before running AMOS and SEM.
66
4. CONFIRMATORY FACTOR ANALYSIS (CFA):
After completing the EFA exploratory factor analysis step, we have the Pattern
Matrix rotation matrix table. Using this rotation matrix result, we start with the
CFA confirmatory factor analysis step, to form the basis for SEM and give the
final result.
Both Hair et al. (2010) and Hu & Bentler (1999) both agree that the thresholds
for accepting the Model Fit index will be different based on sample size,
number of factor groups, number of observed variables,... In addition, due to the
limitation of sample size, it is difficult for the GFI value to reach 0.9. Therefore,
a minimum value of 0.8 is still accepted according to Baumgartner and
Homburg (1995), and Doll, Xia, and Torkzadeh (1994).
According to Hair et al. (2010), Multivariate Data Analysis, 7th edition
indicators considered to evaluate Model Fit of a topic include:
CMIN/df ≤ 2 is good, CMIN/df ≤ 5 is acceptable;
CFI ≥ 0.9 is good, CFI ≥ 0.95 is very good, CFI ≥ 0.8 is
acceptable;
GFI ≥ 0.9 is good, GFI ≥ 0.95 is very good;
RMSEA ≤ 0.08 is good, RMSEA ≤ 0.03 is very good.
Therefore, the team will rely on the assessment of Hair et al. (2010) to check the
index:
On the first run of the data, DPI5 shows that Standardized Loading Estimates is
less than 0.5, so it is removed from the observed variables. But then, we use the
two-way Covariances arrow to connect the pairs with high MI correction, as
shown below. After removing as well as concatenating highly correlated pairs,
the following indexes are obtained:
Table 26: CMIN

Model NPAR CMIN DF P CMIN/DF
Default model 125 1338.136 736 .000 1.818
Saturated model 861 .000 0
Independence
41 7677.167 820 .000 9.362
model
Table 27: RMR, GFI

Model RMR GFI AGFI PGFI
Default model .115 .859 .835 .735
Saturated model .000 1.000
67
Independence
.483 .342 .309 .326
model
Table 28: Baseline Comparisons

NFI RFI IFI TLI
Model CFI
Delta1 rho1 Delta2 rho2
Default model .826 .802 .913 .902 .912
Saturated model 1.000 1.000 1.000
Independence
.000 .000 .000 .000 .000
model
Table 29: RMSEA
RMSE PCLOS
Model LO 90 HI 90
A E
Default model .045 .041 .049 .979
Independence
.145 .142 .148 .000
model
CMIN/DF index is 1,818 (CMIN/DF < 2).

GFI index is 0.859 ( 0.8 < GFI < 0.9).
CFI index is 0.912 (CFI > 0.9).
TLI is 0.902 (TLI > 0.9).
RMSEA index is 0.045 (RMSEA < 0.8).
For the most part, the indicators are good and qualified. However, for the GFI
index, due to the sample size and survey conditions of the group and the authors
Hair et al. (2010) (the number of survey samples of their group is 250), the
results are in the acceptable range.
Therefore, the model meets and meets the implementation requirements as well
as is meaningful to conduct research and is an important step to develop the
SEM model.
68
5. STRUCTURAL EQUATION MODELING (SEM):
After analyzing the CFA, the research team continued to analyze the variables
by SEM to find the relationship between the independent variables and the
intermediate and dependent variables and evaluate the hypothesis for the study.
In this study, the group used the program Amos 24.0, p-value (sig) <= 0.05 with
95% confidence. The results of the analysis are as follows:
Figure 13. Results of Model and Hypotheses Testing
69
Table 30: Summary of the normalized and values
Regression Weights Standardized
Regression
Weights
Estimate P (sig) Estimate
PPR<---TRUST -.156 .020 -.161
PPR<---KNL .039 .357 .052

PPR<---PE .330 .000 .278
PPR<---PREX .401 .000 .400
DPI<---PPR .187 .025 .143
DPI<---PEB -.150 .162 -.118
DPI<---HA .200 .008 .221
DPI<---POLICY -.068 .354 -.059
With the above results we see, when using the standard confidence level of 95
% then the sig of the TRUST variable affecting the intermediate variable PPR is
0.02 < 0.05, so we say that the variable TRUST has an effect on the
intermediate variable PRR and has a standardized regression coefficient of
-0.156 that has a negative effect on the PPR. that is, when TRUST increases by
1, PPR decreases by 1.
The variable KNL does not affect the intermediate variable PPR, the sig of KNL
affecting PPR is 0.357 > 0.05.
The variable PE has an impact on the intermediate variable PPR, the sig PE
effect on PPR is 0.000 < 0.05 and has a standardized regression coefficient of
0.330, which means that when PE increases by 0.330, PPR increases by 0.330.
Similarly, the PREX variable also has an impact on the intermediate variable
PPR with a sig value of 0.000 < 0.05 and a standardized regression coefficient
of 0.401.
70
Among the four variables affecting the dependent variable, there is an
intermediate variable PPR and the independent variable HA has an impact on
the dependent variable DPI, they all have sig less than 0.05, and two
independent variables PEB and POLICY have no impact on the dependent
variable. DPI dependent variable and sig are both greater than 0.05. From the
above results, we conclude that in the 4 independent variables affecting the
intermediate variable PPR, H5 is rejected corresponding to the variable KNL
and has a significant impact on PPR in the order PREX, PE, TRUST. For 4
variables affecting DPI dependent variables, H2 and H8 correspond to PEB and
POLICY variables and have significant effects on DPI in the order HA, PPR.
In the research model, there is an intermediate variable, PPR, to see the

relationship of the variables with each other. The research team has analyzed
the intermediate variable to determine which variables have indirect effects on
the variable. DPI dependence through the intermediate variable PPR. This
analysis relies on a p-value <= 0.05 to conclude for the variables. In this
analysis, the research team only considers 4 variables that have an indirect
impact on the DPI dependent variable through the PPR intermediate variable,
including TRUST, PE, KNL, PREX variables. Analysis results are summarized
in the following table:
Table 31: Indirect and direct effects of variables

Direct Intermediate
Termites impact
S.ES Sig S.ES Sig
TRUST-> PPR-> DPI -0.023 0.018

PEB->DPI -0.118 0.162
HA-> DPI 0.221 0.008
PE-> PPR->DPI 0.040 0.020
KNL-> PPR->DPI 0.007 0.265
POLICY->DPI -0.059 0.354
PREX->PPR->DPI 0.057 0.26
71
PPR->DPI 0.143 0.025
Of the four variables considered, only the KNL variable has no indirect effect
on the dependent variable through intermediate variable PPR and has a sig value
of 0.265 > 0.05. The remaining variables TRUST, PE, PREX, all affect the
dependent variable indirectly through the intermediate variable PPR and all
have sig value less than 0.05, but the PREX variable has a sig value of 0.05 and
that result is possible to temporarily conclude it has an indirect effect on DPI
through PPR. Among the three variables identified to have an indirect effect on
DPI through the intermediate variable PPR, the independent variable PREX
with the most significant impact on DPI has a standardized regression
coefficient of 0.057, followed by PE with a coefficient of 0.05. standardized
regression is 0.040, and finally, the variable TRUST has a significant negative
impact with a standardized regression coefficient of -0.023.
6. TESTING THE RESEARCH HYPOTHESIS:.

After analyzing the SEM, the research team tested 8 hypotheses mentioned in
the research paper.
Test H1: (TRUST) High user trust in the media has a positive impact on the
perception of Facebook and Instagram users. After SEM analysis, the variable
TRUST has an impact on the intermediate variable PPR with the value sig 0.020
< 0.05 and has a negative impact on the variable PPR, therefore, rejecting the
hypothesis H1.
Hypothesis test H2: (PEB), The perceived benefits have a positive effect on the
intention to disclose personal information on Facebook and Instagram.
According to the SEM analysis, the PEB variable has an impact on the DPI
dependent variable with a sig value of 0.162 > 0.05, therefore, the conclusion
that the PEB variable has no impact on the DPI dependent variable, thus
rejecting the hypothesis H2.
Hypothesis test H3:(HA) The intention to disclose personal information on

Facebook and Instagram has increased due to habit. According to the SEM
analysis results, the variable HA affects the dependent variable with a sig value
72
of 0.008 < 0.05 and has the same effect as the DPI variable, therefore, the
conclusion of hypothesis H3 is correct for this study.
Hypothesis test H4: (PE) Personalization positively affects the perception of

Facebook and Instagram users about privacy risks. According to the SEM
analysis results, the variable PE affects the intermediate variable PPR with a sig
value of 0.000 < 0.05 and has a positive effect on the intermediate variable, so
the conclusion of hypothesis H4 is correct for the study.
Hypothesis test H5: (KNL) User perception of privacy risks on Facebook and
Instagram is positively influenced by user knowledge. According to the results
of SEM analysis, the variable KNL affects the intermediate variable PPR with a
sig value of 0.357 > 0.05, concluding that the variable KNL does not affect the
intermediate variable PPR and rejecting the hypothesis H5.
Hypothesis test H6: (PPR) Perception of privacy threats has a negative impact
on intention to disclose personal information. According to the results of SEM
analysis, the intermediate variable PPR affects the dependent variable DPI with
a sig value of 0.025 < 0.05 and has the same or positive effect on the DPI
dependent variable, thus rejecting the hypothesis H6.
Hypothesis test H7: (PREX) Intrusive privacy experience has a great influence
on risk perception. According to the SEM analysis results, the PREX variable
affects the intermediate variable PPR with a sig value of 0.000 < 0.05 and has
the same effect as the intermediate variable PPR, concluding that the hypothesis
H7 is true for the study.
Hypothesis testing H8: (POLICY) The perceived effect of the privacy policy in
reducing the intention to disclose personal information on Facebook and
Instagram. According to the SEM analysis results, the variable POLICY affects
the dependent variable DPI with a sig value of 0.354 > 0.05, concluding that the
variable POLICY does not affect the dependent variable, so rejecting the
hypothesis H8.
73
CHAPTER 5: CONCLUSIONS
1. SUMMARY RESULTS:
The survey of factors influencing risk perception and intention to disclose
personal information began with seven independent variables chosen by the
group after reading and researching related reports. and has been thoroughly
edited through the qualitative process. However, those are also the seven factors
influencing perception and intention that were included in the actual survey
with subjects aged 15 to 29 in Vietnam. The 7-scale Likert scale is used to
assess the level of a specific and most accurate one for the observed variables
measured in the study.
Running data, testing the scale, removing unsatisfactory observed variables, and
observing parameters from Cronbach's Alpha, EFA, CFA, and SEM are all
implemented. The topic's outcome was provided by the team: After receiving
the analysis results for the seven initial hypotheses, the research team rejected
five of them: H1, H2, H5, H6, H8, in which the variables of Hypothesis H2, H5,
H8, respectively, the variables PEB, KNL, POLICY do not affect the dependent
variable and the intermediate variable, while the hypotheses' variables H1, H6,
TRUST, and PPR affect an intermediate variable with the dependent variable.
The variables TRUST, PE, and PREX have a direct impact on individual
privacy risk perception in the order PREX, PE, and TRUST, and thus indirectly
affect the intention to disclose user information. And HA is a factor that has a
direct impact on the user's disclosure intent.
2. MEANING:
The research paper's initial goal is to investigate the factors influencing privacy
risk perception and personal information disclosure; examines the relationship
of perceived privacy risk with the intention to disclose personal information on
Facebook and Instagram, thereby proposing a solution for the study. Based on
the above goal, the research team gathered data on variables thought to affect
74
privacy risk perception and personal information disclosure and analyzed the
SEM model to show the new relationship of the variables together.
Privacy risk perception (PPR) influencing variables include user trust variables
-TRUST, personalization -PE, user knowledge -KNL, and privacy-infringed
experience - PREX. Only the user knowledge variable is not statistically
significant for the study, implying that it has no effect on the perception of
privacy risk in this case study.
There is statistical significance in this study for the remaining three variables,
with the variable user's trust affecting the perceived privacy risk variable with
the standard regression value based on the results of SEM model analysis is [-
0.161]. This figure indicates that if user trust or confidence is high in social
networking sites such as Facebook and Instagram, the perception of user
privacy risk is low; this result is beneficial for the research team to have a fresh
look at consumer behavior. For businesses, it takes a long time to develop a
product that will gain the trust of customers. They must ensure that their
products are as risk-free as possible in order to maintain consumer trust because
when consumers choose to trust, they always expect the product's risk to be as
low as possible or zero.
With a standardized regression coefficient of [0.278], the personalization

variable has an impact on the perceived privacy risk variable; this number is
significant if users' personal information is collected for advertising purposes.
When businesses' products and services are improved, the perception of privacy
risks regarding personal information on Facebook and Instagram increases; this
result indicates that users are very concerned about their personal information
being used maliciously at their own risk.
The final variable that influences privacy risk perception is the experience of
privacy invasion, which has a normalized regression coefficient of [0.400] and
explains whether or not a person has ever experienced a privacy breach. The
facts experienced in practice can have a strong impact on people's perception,
similar to a business if the product launch process leaves a bad impression on
the actual user experience, then the product is considered to have failed in
launching and advertising because it was thought to be bad.
Next, the variables influencing social network users' intention to disclose

personal information - DPI. Direct variables include privacy risk perception
75
-PPR; consumer usage habits -HA; awareness of the benefits provided by social
networks -PEB; and the privacy policies of those social networks -POLICY.
Among the aforementioned variables, there are two that are not statistically
significant for the study, i.e They do not affect the user's intention to disclose
personal information in this research case, namely, the perception of benefits
provided by social networks and the privacy policies of those social networks.
There is statistical significance in this study for the other two variables, the
perceived privacy risk variable based on the results of the SEM model analysis
affects the user's intention to disclose personal information. With a normalized
regression value of [0.143], this number indicates that when users are fully
aware of the risks and consequences of stolen privacy, their intentions to post
information are reduced and their vigilance for future posts is increased. Youn
(2005) discovered that "adolescents are less likely to provide personal
information to a website when they perceive a more serious privacy risk."
Based on the results of the SEM model analysis, the impact on the variable of
user intention to disclose personal information has a standardized regression
value of [0.221] for the variable usage habits of customers. This means that if a
person has a habit of regularly updating images and information on social
networks like Facebook and Instagram, it will continue indefinitely. According
to Beldad et al. (2011), the benefits of information disclosure are not the only
reason people share information, but also the "taste" of the disclosure itself
(p.226). The findings of Strater and Richter (2007), who discovered that some
of their respondents were unsure why they shared information, support the
possibility of a strong influence of usage habits. Others didn't hesitate to provide
personal information when asked because they were accustomed to filling out
forms.
Furthermore, because of the direct impact of the PPR variable, there are three
variables that have an indirect influence on social network users' intention to
disclose personal information: -DPI, which is the user's trust variable -TRUST,
personalization -PE, and privacy-invaded experiences -PREX.
With a standardized regression coefficient of [-0.023], the user trust variable has
an indirect impact on the information disclosure intention variable. If the user's
trust or trust is high for social networking sites Facebook and Instagram, their
perceived privacy risk will be low, affecting their intention to continue sharing
in the future. This finding provides the research team with a fresh perspective
76
on consumer behavior. Businesses must ensure that their products pose the least
amount of risk in order to maintain consumer trust. When consumers choose to
use a product, they can say that they trust and believe that the product meets
their needs and that they intend to use it again in the future. Because TRUST
has a negative effect (-) on PPR and PPR has a positive effect (+) on the DPI
variable, TRUST's effect on DPI has the sign (-).
With a standardized regression coefficient of [0.040] for the personalization

variable, which has an indirect impact on the information disclosure intention
variable, this number indicates that the collection of users' personal information
for advertising purposes increases as the number of products and services
offered by businesses increases, as does the perception of privacy risks
regarding personal information on Facebook. This will have an impact on their
willingness to use social networking sites in the future. Because PE has a
positive (+) effect on PPR and PPR has a positive (+) effect on the DPI variable,
the effect of PE on DPI has a (+) sign.
With a standardized regression coefficient of [0.057], PREX has an indirect

effect on the variable of intention to disclose information for those who have
had their privacy violated. This number is significant in explaining the behavior
of a person who has experienced a privacy violation, has a very high awareness
of privacy risks in the future and is wary of shared information; this result is
very meaningful for the person. It has a strong impact on people's perceptions,
behaviors, and attitudes, just like a business, the product launch process leaves a
bad impression in the process of using its users, they will not continue to choose
and trust the product, and the company's brand will leave a negative impression
in the eyes of consumers. Because PREX has a positive effect (+) on PPR and
PPR has a positive effect (+) on the DPI variable, PREX's effect on DPI has a
(+) sign.
Users will gain more knowledge and a better understanding of their own
personal information as a result of the above results, increasing their vigilance.
At the same time, it provides Facebook and Instagram security administrators
with an overview of the situation, allowing them to develop solutions that help
protect customers' personal information while minimizing risk for user privacy.
Furthermore, based on the survey results, we make recommendations to social
networking site users and social network service providers to help them better
understand the risks associated with sharing personal information.
77
3. LIMITATION OF THEME AND FUTURE RESEARCH
DIRECTIONS
This study provides specific results and contributions that will assist social
networking site owners and users in understanding the factors that influence
perceived privacy risks and intentions to disclose personal information. The
multiplier set by the user. However, there are some limitations to this study:
Firstly, due to research constraints such as time, cost, and the seriousness of the
Covid-19 epidemic, the team only focused on interviewing and surveying users
on a small scale, as well as sending online questionnaires. However, if the scope
of interviews and surveys can be broadened, the research will yield more
general results. As a result, future studies must be carefully invested in and
directly investigated for more positive and practical outcomes.
Secondly, the research method used in this study is a convenient sampling
method, with a small sample size, so the research results will not be as profound
and provide the desired results as a larger sampling method. As a result, it is
prudent to invest in and expand the survey space in order to ensure the validity
and generality of the results.
Thirdly, it was difficult to make an appointment to meet with an expert during
the interview process, so the team could only interview via online message;
additionally, the team lacked in-depth knowledge, so asking questions did not
work. It is not necessary to thoroughly investigate the impact of factors
influencing user perception when promoting topics and goals. As a result,
improving skills as well as learning, listening, and understanding will aid in
developing a deeper understanding of the content as well as the observed
variables of the lesson.
Fourthly, because the team learned the software that runs the data through self-
study and watching video tutorials on YouTube, the team spends a significant
amount of time processing the data tables in SPSS and AMOS. However,
reading the results is also a shortcoming of the article because the team only
focuses on the samples available in documents and videos, resulting in a
shallow and incomplete conclusion.
Finally, the study only surveys seven factors (7 observed variables) that
influence the perception of privacy risks and the intention to disclose personal
information on social media. However, the final results show that only four
factors: HA, PE, PREX, and TRUST, have a direct or indirect impact on the
78
user's perception and behavior. As a result, the results show that the factor
observation and modeling are not complete and professional, and the impact
rate on factors is low. Future research should focus on the effect of demographic
factors on user perception and behavior, as well as learning more about other
factors that influence user perception and behavior. A longitudinal survey is
needed to investigate the effects of perceived control over personal information,
subjective norms, and privacy computational models on information disclosure
of a private message.
4. SOLUTION
It is now normal practice to update information on social networking sites or to
share photographs and content relating to persons. It has put users at danger of
being sold personal information for nefarious purposes, getting defrauded or
losing money as a result of disclosing too much personal information, and so
on. Users are always exposed to the threats that social networking sites like
Facebook and Instagram pose, as well as the benefits that they provide. We can
offer the following recommendations based on the study's findings:
* With social network participants:

Given the privacy issues, we believe that today's social media users should learn
more about social networking site privacy policies and the hazards that come
with them. They do so at their own risk with the information they disclose. Then
there's the question of how to safeguard and mitigate those dangers in the future.
Today, there are several incidents of fraud involving the fabrication of personal
information on social media, which has a significant impact on the exposed
person's mental and physical health, as well as that of their family. As a result,
we must responsibly share information in order to meet our personal
requirements while also protecting ourselves, our friends, and family.
The advice given is to disclose as little information as possible, or to disclose
information only when it is really required and reliable. Simultaneously, crucial
personal information such as date of birth, phone number, home address,
family, workplace address, etc. must be kept private.
Users should also note that many applications can save moving location as well
as many applications use artificial intelligence to know activities such as
searching, check-in, shopping, even eavesdropping on people's use. Therefore,
users need to use copyrighted software so as not to lose personal information
79
and use applications from official sources provided by reliable units. Do not
visit unknown websites or access links on Facebook and Instagram.
Users should also use specialized anti-virus and anti-malware software, which
should be updated regularly to avoid losing important information in personal
devices connected to the Internet.
Before there are effective sanctions to protect users against online attacks and
scams, people need to minimize the provision and sharing of personal
information to ensure their own safety and the people around them.
In this study, we discovered that previous privacy-invading events have the
greatest impact on users' perceptions of danger and their willingness to utilize
social media to share information. As a result, giving evidence of instances
involving the risk of releasing personal information will serve as a warning to
users about the significance and power of information published on social
media platforms.
In addition, photos and videos posted on social networks, seemingly without
revealing personal information, are actually becoming data that are analyzed,
aggregated, and given advertising suggestions according to needs. user demand.
Therefore, users need to consider carefully before posting any information on
social networks.
Moreover, network security officers must learn from the events in order to
improve cybercrime prevention and network security so that consumers can feel
more secure when using social media. Simultaneously, propagandists are urging
and reminding people to be more conscious of the importance of preserving
their own and others' personal information.
Furthermore, putting complete faith in social network privacy policies is not
advisable and potentially dangerous. Because, in reality, social media's existing
security strategy is riddled with flaws and threats. As a result, it's impossible to
foresee whether they'll offer us benefits or risks, and we shouldn't put too much
faith in them.
* With social network service providers:

Social network service providers can customize services, improve their core
competitiveness and continue to grow by collecting complete user information.
This study provides new ideas for the platform to better understand user
disclosure behaviors and thus more information about users.
Providing users with precisely personalized services, social platforms can
improve the benefits perceived by users and promote user disclosure behavior.
80
This study indicates that by improving the cybersecurity environment,
information disclosure can be promoted by reducing perceived risk. Therefore,
social networking platforms should provide high-quality privacy policies and
place them in prominent positions to reduce perceived risk. Because of the
higher perceived risk by users in the mandatory provision, platforms should try
to avoid collecting large amounts of sensitive information during the
registration process and should only ask users for information instead of
believing the security. Registered users will find it easier to disclose more
information because they have decided that the risk is low.
81
REFERENCES
Albert L. Ball, Michelle M. Ramim,Yair Levy (2015). Examining users’
personal information sharing awareness, habits, and practices in social
networking sites and e-learning systemshttps://bitly.com.vn/e0zfya
Alemany, J, Del Val, E. and Garcia-Fornes, A.M. (2021),“Who should I grant
access to my post?”: Identifying the most suitable privacy decisions on online
social networks.https://bitly.com.vn/xizk05
Alisa Frik, Alexia Gaudeul (2020). A measure of the implicit value of privacy
under risk.https://bitly.com.vn/cwtzdk
Ashraf Sharif, Saira Hanif Soroya, Shakil Ahmad and Khalid Mahmood (2021).
Antecedents of Self-Disclosure on Social Networking Sites (SNSs): A Study of
Facebook Users.https://bitly.com.vn/iqhb7y
Aungkana Jattamart, Adisorn Leelasantitham (2020). Perspectives to social
media usage of depressed patients and caregivers affecting to change the health
behavior of patients in terms of information and perceived privacy risks.
Baumgartner, H., Homburg, C.: Applications of Structural Equation Modeling
in Marketing and Consumer Research: a review. International Journal of
Research in Marketing 13(2), 139-161 (1996).
Bauer, 1960. Theory of Perceived Risk.
Beldad et al. (2011), the benefits of information disclosure are not the only
reason people share information, but also the "taste" of the disclosure itself
(p.226).
Boyd, 2009. Why teens have a strong presence and visibility on online social
networks.
Byoungsoo Kim; Daekil Kim (2020). Understanding the Key Antecedents of
Users’ Disclosing Behaviors on Social Networking Sites: The Privacy
Paradox.https://bitly.com.vn/sp1vwy
Chang, Y., Wong, S. F., Libaque-Saenz, C. F., & Lee, H. (2018). The role of
privacy policy on consumers' perceived privacy. Government Information
Quarterly, 35(3), 1–15.
Chantal Mutimukwe, Ella Kolkowska, Åke Grönlunda(2019). Information
privacy in e-service: Effect of organizational privacy assurances on individual
privacy concerns, perceptions, trust and self-disclosure
behavior.https://bitly.com.vn/u70v06
Doll, W.J., Xia, W., Torkzadeh, G.: A confirmatory factor analysis of the end-
user computing satisfaction instrument, MIS Quarterly 18(4), 357–369 (1994).
82
E.g Grant, 2005, 2006; Youn, 2005, 2008; Moscardelli & Divine, 2007.
Researchers started to empirically research adolescents and their attitudes
toward online privacy concerns.
E.g. Livingstone, 2008; Lenhart & Madden, 2007; Romer, 2006. The
information disclosure of young adolescents on these services has intensified
worries about loss of privacy.
George Oppong Appiagyei Ampong, Aseda Mensah, Adolph Sedem Yaw Adu,
John Agyekum Addae, Osaretin Kayode Omoregie and Kwame Simpe Ofori
(2018). Examining Self-Disclosure on Social Networking Sites: A Flow Theory
and Privacy Perspective.https://bitly.com.vn/rlypab
Habit develops in the absence of experience or thought (Bargh, 1994; Nosek,
Hawkins, & Frazier, 2011), which may be influenced by environmental
perceptions and performers rather than conscious awareness (Verplanken,
Myrbakk, & Rudi, 2005).
Habit is described as a set of orchestrated behaviors that are activated by
specific cues and lead to a specific outcome (Verlinken & Arts, 2006).
Hongliang Chen, David Atkin(2020). Understanding third-person perception
about Internet privacy risks.https://bitly.com.vn/qb5xqp
Hair et al. (2010), Multivariate Data Analysis, 7th edition.
Juan Shi, Kin Keung Lai, Gang Chen (2020). Individual Retweeting Behavior
on Social Networking Sites: A Study on Individual Information Disseminating
Behavior on Social Networking Sites.https://bitly.com.vn/1b1374
Kai Li, Xiaowen Wang, Kunrong Li, Jianguo Che(2016). Information privacy
disclosure on social network sites.https://bitly.com.vn/xxae8a
Kai Lia , Liangqi Chenga , Ching-I Teng(2020). Voluntary sharing and
mandatory provision: Private information disclosure on social networking
sites.https://bitly.com.vn/j2ngep
Krasnova, H., Spiekermann, S., Koroleva, K., & Hildebrand, T. (2010). Online
social networks: why we disclose. Journal of Information Technology, p.109-
125.
Leon Caesarius (2015). Disclosing personal information to social networking
site providers The role of trust, risk and perceived
benefits.https://bitly.com.vn/gr5wdu
Mashael Aljohani, Alastair Nisbet, Kelly Blincoe (2016). A survey of social
media users' privacy settings & information
disclosure.https://bitly.com.vn/jk1tjx
83
Nguyen Bich (24/7/2019).Lot, lo thong tin ca nhan tren mang Internet: Nhung
nguy co kho luong. https://bitly.com.vn/zllau6
Nina Michael Idou, Milena Maticevski (2016). Consumers' ethical perceptions
of social media analytics practices: Risks, benefits and potential
outcomes.https://bitly.com.vn/7mw9jk
Office of the Assistant for Administration & Managements. Secretary Privacy
Impact Assessment - OPA - Social Media Services. https://bitly.com.vn/bz2m4i
P. Krubhala, P.Niranjana, G.Sindhu Priya (2015).Online Social Network - A
Threat to Privacy and Security of Human Society.https://bitly.com.vn/g159jg
Paul van Schaik, Jurjen Jansen, Joseph Onibokun, Jean Camp, Petko Kusev
(2017). Security and privacy in online social networking: Risk perceptions and
precautionary behaviour.https://bitly.com.vn/5qzaye
PGS.TS. Nguyen Thi Nhung (2020). Protecting the rights to personal data and
information in the current social networking environment in
Vietnam.https://bitly.com.vn/ee3dwz
Phillip Kotler, 2001. Consumer behavior theory.
Population Pyramids of the World from 1950 to 2100. Population of Viet Nam
2019.https://www.populationpyramid.net/viet-nam/2019/
Power and Trope (2006), Users' behaviors may influence their activities.
Rami Baazeem, Alaa Qaffas (2020). The relationship between user religiosity
and preserved privacy in the context of social media and cybersecurity.
RezaMousavi, Rui Chen, Dan J.Kim, Kuanchin Chen (2020). Effectiveness of
privacy assurance mechanisms in users' privacy protection on social
networking sites from the perspective of protection motivation theory.
Ruud H.G. Koehorst, Dr. A. Beldad and Dr. J.M. Gutteling (2013). Personal
Information Disclosure on Online Social Networks.https://bit.ly/3avUlwO
Santor, Messervey & Kusumakar, 2000. For teens the need to be a part of a
social group and to be popular are important parts of their lives.
The concept of habit is "the degree to which people appear to execute activities
unconsciously as a result of learning" (Limayem, Hirt, & Cheung, 2007, p.
709).
The New Rules on Social Media, Privacy and Data
Protection.https://bitly.com.vn/8y0c5q
Tuong Quan (25/6/2020). Dung lo la viec bao ve thong tin ca
nhan.https://bitly.com.vn/1v4xps
84
Verplanken, B., & Orbell, S. (2003). Reflections on Past Behavior: A Self‐
Report Index of Habit Strength1. Journal of Applied Social Psychology, 33(6),
p.1313-1330.https://bitly.com.vn/o5hm0l
Youn (2005) discovered that "adolescents are less likely to provide personal
information to a website when they perceive a more serious privacy risk."
Xu, Dinev, Smith & Hart, (2011a). Information privacy concern: Linking
individual perceptions with institutional privacy assurances. Journal of the
Association for Information Systems. https://bitly.com.vn/j2ngep
Xu, H., Dinev, T., Smith, J., & Hart, P. (2011). Information privacy concerns:
Linking individual perceptions with institutional privacy assurances. Journal of
the Association for Information Systems, 12(12), 798–824.
Xu, Luo, Carroll & Rosson, 2011. The personalization privacy paradox: An
exploratory study of the decision making process for location-aware marketing.
Decision support Systems.https://bitly.com.vn/j2ngep
Link Drive: https://drive.google.com/drive/folders/12xP-

3m_38R2dg1Gx9sSELLOn_WutTxqX?
fbclid=IwAR2YoNMV_y_q2UvmTznxe13VYiP0u2htxDEkNeP4KRwvD6ZM
Pp8Kk3t54yU
85
APPENDIX A: QUALIFICATION QUESTION LIST FOR
SPECIALISTS AND FOCUS GROUPS
Interview questions for in-depth interview:
Question 1: Based on your experience posting personal information on social
media, do you believe you can or should not post your personal information and
activities on social media? And how can the data be disseminated? Should
details be selected and screened before being shared on social media platforms
such as Facebook or Instagram? (Objectives 1)
Question 2: How do you think existing consumers consider privacy threats and
the sharing of sensitive information? (Objectives 1)
Question 3: Which causes, in your view, have a strong influence on
expectations of privacy risks as well as the behavior of disclosing personal
information? (Objectives 2)
Question 4: Do you have any solutions or recommendations for safeguarding
sensitive information and privacy on social networking sites? (Objectives 3)
Interview questions for focus group:

Question 1: What do you think about the privacy risks and the sharing of
sensitive information on social media? (Objectives 1)
Question 2: What factors do you believe influence people's views of privacy
threats and their willingness to share personal information on social media?
Why is this so? (Objectives 2)
Question 3: Do you have faith in Facebook and Instagram's security? If so, what
is the percentage? (Objectives 2)
Question 4: Do you ever share your personal knowledge, either publicly or
privately? Why is this so? (Objectives 2)
Question 5: How do you believe consumer trust in a network can impact
privacy risk awareness and personal information disclosure? (Objectives 2)
Question 6: What are you hoping to do by sharing personal details on social
networking sites (such as Facebook and Instagram)? Or what advantages do you
get by sharing personal information on social media? (Objectives 2)
Question 7: How does your personality influence your understanding of privacy
threats and social media disclosure? (Objectives 2)
Question 8: What are your thoughts on the relationship between privacy risk
expectations and personal information disclosure? (Objectives 1)
Question 9: How can you cover yourself by revealing personal information on
social networking sites (such as Facebook and Instagram)? (Objectives 3)
86
APPENDIX B: QUALITATIVE RESPONSE LIST
In-depth interview:
STT Full name Job Age Perform
er
1 Tran Thanh Tung Working at FOX SPORT 31 N.Hang
Focus group:
STT Full name Job Age Perfor
mer
1 Nguyen Dang Minh Quan Student at Ho Chi Minh City 20 T.Nhi
University of Technology
2 Tran Anh Minh Student at University of 25 T.Nhi
Information Technology
3 Ro Cham Bien Student at Ho Chi Minh City 20 P.Be
University of Medicine and
Pharmacy
4 Nguyen Thi Anh Thu Student at University of Foreign 20 P.Be
Language Studies
5 Le Hoang An Student at Ho Chi Minh City 20 T.Hieu
University of Economics
6 Nguyen Thao Nguyen Student at Ho Chi Minh City 20 T.Hieu
University of Economics
7 Tran Nguyen Nhu Quynh Student at Thanh Nhan High 16 N.Duye
School n
87
APPENDIX C: QUANTITATIVE QUESTIONNAIRE FOR
SURVEY
UNIVERSITY OF ECONOMICS HO CHI MINH CITY
SCHOOL OF INTERNATIONAL BUSINESS - MARKETING
MARKETING RESEARCH
TOPIC:
PRIVACY RISK AWARENESS AND INTENTIONS OF DISCLOSURE
OF PERSONAL INFORMATION FACEBOOK AND INSTAGRAM.
ID:
Hi everybody.
We are a group of students from the Faculty of International Business -
Marketing, University of Economics Ho Chi Minh City (UEH). Currently using
social networking sites is very popular for each of us and especially Facebook,
Instagram, ... And whether everyone can be aware of the potential risks. behind
share buttons or after posts on a personal page. So our team conducted a survey
on the topic "Privacy risk awareness and intent to disclose personal information
of social media users Facebook and Instagram" for the study. Marketing, as well
as being able to share with users solutions, or suggestions to protect themselves
against the strong development of the internet in today's society.
Hope you spend some valuable time helping our team complete the questions in
this survey to support the research process. And the team pledges that all of
your information and opinions will be kept confidential and used for academic
purposes only.
Sincerely thank!
88
No Measurement items
PART 1: DEMOGRAPHIC QUESTIONS
1. What is your gender?
Male
Female
Others
2. What is your age?
From 15 to 19
From 20 to 24
From 25 to 29
3. What is your profession?

Civil servants / public employees
Enterprise owners / leaders
Employees
Self-Sales Services
Skilled Workers
Small Business
Simple labor
Retirement
Students - Student
Not working (Housewife / Unemployed)
4. Do you use Facebook and Instagram social networks?
Yes
No
5. Frequency of using social networks above: (hour / day)
Less than 1 hour
2-3 hours
3-5 hours
More than 5 hours
89
6. How long do you normally use a Facebook or Instagram
account?
Less than 1 year
From 2-3 years
From 3-5 years
More than 5 years
7. How many friends do you have on Facebook or Instagram?

Less than 100
About 100-500
About 500-1000
More than 1000
PART 2: QUESTIONNAIRE TO STUDY ON VARIABLES.

Note: In this section we ask questions that are measured on a scale of 1-7,
corresponding to the answer from Totally disagree- Totally agree, Totally
disagree- Fully understand Clear, Never- Many times. Please put an X in the
box you think best fits.
TRUST
1-7(Completely disagree- completely agree) 1 2 3 4 5 6 7
1 I trust that Facebook and Instagram have the
expertise to handle my personal information.
2 Facebook and Instagram have good intentions for
my personal information.
3 I trust that my Facebook and Instagram-friends
have the expertise to not jeopardize my personal
information.
90
don’t do anything with my personal information I
would not approve of
keep my preferences and desires about my
personal information in mind.
6 I trust that Facebook and Instagram protect my
personal information against companies and
advertisers that want to abuse my information.
7 I trust that Facebook and Instagram check if
everybody is playing by the rules.
PERCEIVED BENEFITS
8 Facebook and Instagram are useful to exchange
personal information with your friends.
9 Thanks to sharing personal information on
Facebook and Instagram, I get to know people
better.
10 Facebook and Instagram are useful for me to
monitor what others share about themselves.
11 Sharing personal information on Facebook and
Instagram is fun.
12 On Facebook and Instagram I have more courage
in sharing personal information compared to other
situations.
13 By sharing personal information on Facebook and

Instagram, I get more popular with my Facebook-
friends.
14 I share personal information via Facebook and
Instagram because it’s better than the alternatives.
15 By sharing personal information on Facebook and
91
Instagram, I can make a good impression on my
Facebook-friends.
HABIT
16 Sharing personal information is a habit to me.
17 I sometimes share personal information without
thinking about it.
18 I sometimes share personal information because it
is hard for me not to share it with others
19 If I see something interesting, the first thing that
comes into mind is to share it on Facebook.
PERSONALIZATION
20 Facebook and Instagram can provide me with
personalized advertising and transactions.
21 Facebook and Instagram provide me with a
corresponding push message based on my
preferences or personal interests.
22 Facebook and Instagram provide me with ads I
like.
KNOWLEDGE
1-7(Totally incomprehensible- Completely 1 2 3 4 5 6 7
understood)
23 Perceived understandings about virus attacks.
24 Perceived understanding about hacker attacks.
25 Perceived understandings about identification
theft.
26 Perceived understandings about credit card theft.
92
27 Perceived understandings about privacy invasion.
28 Perceived understanding about online insults.
PERCEPTION OF PRIVACY RISKS

29 Facebook and Instagram as a company is a danger
for the safety of my personal information.
30 I’m afraid that Facebook and Instagram sell my
personal information to others.
31 I’m afraid that Facebook and Instagram secretly
use my personal information for purposes I don’t
agree with.
32 Friends-of-friends or companies on Facebook and
Instagram are a danger to my personal information.
33 Hackers are a danger for my personal information
on Facebook and Instagram.
34 I’m afraid my Facebook and Instagram-friends get
a wrong impression of me because of the personal
information I’ve shared on Facebook.
35 My Facebook and Instagram-friends are a danger
to the safety of my personal information.
36 I’m afraid my personal information can be used by
my Facebook and Instagram friends to bully me
with.
37 I don’t care about the risks of sharing personal
information.
38 I’m afraid that I unintentionally shared personal
information because I made a mistake.
39 There are dangers to sharing personal information
93
that I’m not aware of.
PRIVACY INVASION EXPERIENCE
1-7 (Never- always) 1 2 3 4 5 6 7

40 How many times have you felt that your privacy
has been violated?
41 In the past year, how many times have you heard
about the indiscriminate use of private information
on the Internet?
42 How much have you heard or read during the last
year about the use and potential misuse of
information privacy about consumers?
43 How often have you personally been a victim of
what you felt was an invasion of privacy?
PRIVACY POLICY
44 I feel confident that privacy statements on
Facebook and Instagram reflect their commitments
to protect my personal information.
45 With privacy statements, on Facebook and

Instagram I believe that my personal information
will be kept private and confidential.
46 I believe that privacy statements on Facebook and
Instagram are an effective way to demonstrate
their commitments to privacy.
INTENTION TO DISCLOSURE PERSONAL INFORMATION
1-7 (Never- always) 1 2 3 4 5 6 7
47 I filled in what town I live in.
94
48 I filled in where I work.
49 I filled in who my family members are.
50 I am filled in if I have a relationship.
51 I filled in my phone number.
52 I often share my opinion on Facebook and
Instagram.
53 I often share on Facebook and Instagram where I
am at that moment.
We sincerely thank you for taking the time to complete the survey for us. And
hopefully the questions of the group topic can partly help people orient and
protect themselves.
Once again sincerely thank you, and good day everyone.
95
APPENDIX D: CFA DIAGRAM
96
97
APPENDIX E: SECONDARY DATA SOURCES
Link: https://bitly.com.vn/xlmz8e
98
PERFORMANCE ASSESSMENT TABLE:
Members Exercise 1 Exercise 2 Exercise 3 Exercise 4 Report
Nguyễn Bình Thảo 100% 100% 100% 100% 100%
Nhi
Bế Nguyễn Hằng 100% 100% 100% 90% 100%
Nguyễn Thị Ngọc 100% 100% 100% 100% 100%

Duyên
Puih Bé 100% 100% 100% 100% 100%
Vũ Thanh Hiếu 100% 100% 100% 95% 100%
MỤC LỤC BẢNG
99

Ncmaren Ib5clck45 Final-Report n06

Uploaded by

Copyright:

Available Formats

Ncmaren Ib5clck45 Final-Report n06

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ncmaren Ib5clck45 Final-Report n06

Uploaded by

Copyright:

Available Formats

UNIVERSITY OF ECONOMICS HO CHI MINH CITY

SCHOOL OF INTERNATIONAL BUSINESS - MARKETING

Supervior: PhD. DINH TIEN MINH

HO CHI MINH CITY, 2021.

SUPERVISOR: PHD. DINH TIEN MINH

NAME STUDENTS EMAIL

Our team sincerely thank you!

“ Nobody is literally forced to join an online social network, and most

2. THE REASON TO CHOOSE THE TOPIC:

3. THE OBJECT OF STUDY:

4. THE OBJECT OF SURVEY:

6. SCOPE OF THE STUDY:

1.1. TPR – Theory of Perceived Risk (Bauer, 1960)

Risk perception theory (TPR)

2.1. Popular sources:

Figure 1. Model of reference for research

*Voluntary sharing and mandatory provision: Private information

Figure 2. Model of reference for research

- The results of this study:

Figure 3.Model of reference for research

2.1.4. Social Media:

*Disclosure of personal information on the Internet:

2.2. Scholarly sources:

*Understanding the Key Antecedents of Users’ Disclosing Behaviors on Social

- Results: Consistent with expectations, users’ intention to disclose personal

*Examining Self-Disclosure on Social Networking Sites: A Flow Theory and

*Antecedents of Self-Disclosure on Social Networking Sites (SNSs): A Study of

*A survey of social media users privacy settings & information disclosure

*Personal Information Disclosure on Online Social Networks

Figure 6. Model of reference for research

*Information privacy disclosure on social network sites

Figure 7. Model of reference for research

*Information privacy in e-service: Effect of organizational privacy assurances

Figure 8. Model of reference for research

2.3. Government sources:

*Privacy Impact Assessment - OPA - Social Media Services

*Disclosing personal information to social networking site providers The role

Figure 9. Model of reference for research

*Examining users’ personal information sharing awareness, habits, and

*Online Social Network - A Threat to Privacy and Security of Human Society

3.2. Effect of perceived benefits on Facebook and Instagram user's privacy

3.3. Effect of habit on Facebook and Instagram user's privacy risk

3.4. Effect of personality on Facebook and Instagram user's privacy risk

3.5. Effect of knowledge on Facebook and Instagram user's privacy risk

3.6. Effect of perception of privacy risk on personal information disclosure

4. CONCLUSION THEORY FOR GROUP TOPIC:

Figure 11. Research model

2.1. How to do it:

2.2. Results of qualitative research

2.2.1. Research model after qualitative research:

A. Effect of privacy invasion experience on risk perceptions:

B. Effect of privacy policy on intention to disclose personal information:

2.2.2. Scale adjustment for the topic:

B. Scale of “Perception of benefits”

F. Scale “Privacy risk perception”

G. Scale of “Privacy Intrusive Experience”

I. Scale of “Intent to disclose personal information”

Table 1: Results of qualitative research and scale

No Measurement items Items