=Paper=
{{Paper
|id=Vol-2348/short08
|storemode=property
|title=Measuring the Impact of Privacy Concerns, Perceived Risk and Trust on the Continuance Intention of Facebook Users by Means of PLS-SEM
|pdfUrl=https://ceur-ws.org/Vol-2348/short08.pdf
|volume=Vol-2348
|authors=Dilek Taşkın,Çağatan Taşkın
|dblpUrl=https://dblp.org/rec/conf/cerc/TaskinT19
}}
==Measuring the Impact of Privacy Concerns, Perceived Risk and Trust on the Continuance Intention of Facebook Users by Means of PLS-SEM==
https://ceur-ws.org/Vol-2348/short08.pdf
Business and Society
Measuring the Impact of Privacy Concerns, Perceived
Risk and Trust on the Continuance Intention of Facebook
Users by means of PLS-SEM
Dilek Taşkın1 and Çağatan Taşkın2
1 Bursa Uludağ University, Bursa, Turkey
2 Bursa Uludağ University, Bursa, Turkey
dilektaskin@uludag.edu.tr
ctaskin@uludag.edu.tr
Abstract. Facebook became one of the most popular ways of online social inter-
action. Many social networking sites like Facebook focus on their market, in
other words, users. That’s why, antecedents of continuance intention of Facebook
users should be investigated in order to develop more efficient marketing strate-
gies [1]. According to the literature, privacy concerns, perceived risk and trust
are the most important antecedents of continuance intention. The aim of this
study is to examine the influence of privacy concerns, perceived risk and trust on
the continuance intention of Facebook users in Bursa city of Turkey. Data were
collected via an online questionnaire. A total of 241 questionnaires were used for
the analysis. According to the results, privacy concerns antecedent was found to
have a statistically negative significant impact on trust and positive significant
impact on perceived risk. Trust antecedent was also found to have a statistically
significant positive impact on continuance intention of Facebook users.
Keywords: Privacy concerns, perceived risk, trust, continuance intention, Face-
book, Bursa, Turkey.
1 Introduction
In recent years, social networks especially Facebook attracted a great number of people
all over the world [2]. Facebook became one of the most popular ways of online social
interaction. Many social networking sites like Facebook focus on their users. The users
of social networks are so important for the long-term success of them [3]. That’s why
antecedents of continuance intention of Facebook users should be investigated in order
to develop more efficient marketing strategies [1].
In the literature, there are studies that aim to understand the relations among continu-
ance intention and its antecedents for social media platforms. According to Wang et al.
(2016), trust and risk were found to have significant effects on individual behavior to-
ward social media platforms but that trust had a stronger effect [4]. Another study com-
351
Business and Society
2
pared two social media platforms (Facebook and LinkedIn) to understand factors af-
fecting users’ trust on social media platforms. According to the findings of the study, it
was found that users’ trust on social media platforms was mainly influenced by effort
expectancy, social influence and perceived risk. Besides, it was found that trust had a
significant impact on continuance intention [2]. In Tan et al. (2012)’s study, it was
aimed to understand the impact of users’ privacy concerns on their acceptance of social
media platforms [5]. There is also another study that proposed a research model to in-
vestigate individuals’ social media platform usage facilitators and inhibitors from the
perspective of privacy concerns [6]. In addition, there are various researches that aim
to understand the antecedents of continuance intention in photo-sharing context [7] [8]
[9] [10]. According to the literature, privacy concerns, perceived risk and trust are the
most important antecedents of continuance intention. Especially, privacy concerns is a
very current topic for the social media market. Thus, it is aimed to examine the influ-
ence of privacy concerns, perceived risk and trust on the continuance intention of Fa-
cebook users in this study.
2 Methodology
2.1 Research Sample and Method
The research was conducted on Facebook users in Bursa city of Turkey. Data were
collected via an online questionnaire. Convenience sampling method was used in the
research. Data were collected in the months of June, July and August of 2018. 275
questionnaires were collected from respondents. 34 of them were excluded as they were
not complete based on the initial screening. A total of 241 questionnaires were used for
the analysis. The antecedents of the model such as privacy concerns, perceived risk,
trust, effort expectancy, social influence and continuance intention were measured by
the items based on the related literature [2] [3] [11]. Smart PLS 3.0 and IBM SPSS 21.0
were used in order to analyze the data. PLS-SEM was used to test the influence of
privacy concerns, perceived risk and trust on the continuance intention of Facebook
users.
The popularity of structural equation modeling (SEM) has grown out of the need to test
complete theories and concepts. Much of SEM’s success can be attributed to the
method’s ability to evaluate the measurement of latent variables, while also testing re-
lationships between latent variables. Although the initial application of this method em-
braced a covariance-based approach (CB-SEM), researchers also have the option of
choosing the variance-based partial least squares technique (PLS-SEM). While CB-
SEM is the more popular method, PLS-SEM has recently received considerable atten-
tion in a variety of disciplines including marketing, strategic management, management
information systems, operations management, and accounting. Much of the increased
usage of PLS-SEM can be credited to the method’s ability to handle problematic mod-
eling issues that routinely occur in the social sciences such as unusual data characteris-
tics (e.g. nonnormal data) and highly complex models [12].
352
Business and Society
3
2.2 Profile of Respondents
Demographic characteristics of the respondents were given in Table 1. As seen, 110 of
the respondents were male and 131 of the respondents were female. Most of the re-
spondents (44.4%) were between 21-30 ages, 34.0% of the respondents were between
31-40 ages and 13.7% of the respondents were between 41-50 ages. Most of the re-
spondents had an undergraduate degree or were more educated (92.6%). Single Face-
book users were slightly more than married users. Most of the respondents had a Face-
book experience for 4 and more years. 56.4% of the respondents spent 60 minutes or
less time on Facebook, while 43.6% spent more than 60 minutes time on Facebook per
day.
Table 1. Profile of Respondents.
Demographics Frequency % Demographics Frequency %
Gender Education
Male 110 45,6 Primary school 3 1,2
Female 131 54,4 High school 15 6,2
Age Undergraduate 106 44,1
<=20 2 0,8 MSc 97 40,2
21-30 107 44,4 PhD 20 8,3
31-40 82 34.0 Marital Status
41-50 33 13,7 Single 133 55,2
>50 17 7,1 Married 108 44,8
Time spent on
Facebook expe-
Facebook per
rience
day
60 minutes or 136 56,4
<1 year 5 2,1
less
More than 60 105 43,6
1-2 years 2 0,8
minutes
2-3 years 8 3,3
3-4 years 9 3,7
4 years and more 217 90,1
2.3 Research Model and Hypothesis
The research model is shown in Fig. 1. As it can be seen, the research model includes
the variables, which are; “effort expectancy”, “social influence”, “privacy concerns”,
“perceived risk” “trust” and “continuance intention” and the relationships among them.
353
Business and Society
4
Fig. 1. Research Model
The hypotheses of the research are:
H1: “Effort expectancy” positively influences “trust”.
H2: “Social influence” positively influences “trust”.
H3: “Effort expectancy” positively influences “continuance intention”.
H4: “Social influence” positively influences “continuance intention”.
H5:“Privacy concerns” negatively influences “trust“.
H6:“Perceived risk” negatively influences “trust“.
H7:“Perceived risk” negatively influences “continuance intention“.
H8:“Privacy concerns” positively influences “perceived risk“.
H9:“Trust” positively influences “continuance intention“.
2.4 Construct Reliability and Validity
Table 2 shows the results of construct reliability and validity. The AVE (Average Var-
iance Extracted) values of the structure must be 0,50 or more for the validity of latent
structures [13]. Results show that the AVE values for effort expectancy, social influ-
ence, trust, continuance intention, privacy concerns and perceived risk are 0.60, 0.55,
0.63, 0.74, 0.76 and 0.56, respectively. As the result of the analysis carried out with
Smart PLS and PLS estimation method, the composite reliability value is given. Com-
posite Reliability value should be 0.70 or above [14].
354
Business and Society
5
Table 2. Construct Reliability and Validity
Construct Cronbach’s Alpha Rho’A CR AVE
Effort Expectancy 0.84 0.89 0.88 0.60
Social Influence 0.80 0.91 0.85 0.55
Trust 0.80 0.85 0.87 0.63
Continuance Inten- 0.88 0.89 0.91 0.74
tion
Privacy Concerns 0.89 0.90 0.92 0.76
Perceived Risk 0.77 0.94 0.83 0.56
2.5 Testing the Research Model by Using PLS
Table 3 shows the results of hypothesis tests and structural relationships. The path co-
efficient of trust on continuance intention is 0.44, the path coefficient of social influence
on trust is 0.41 and the path coefficient of privacy concerns on trust is -0.27. On the
other hand, the path coefficients of perceived risk on trust and continuance intention
are -0.09 and -0.007, respectively.
Table 3. Results of Hypothesis Tests and Structural Relationships
Hypothesis Path Coefficient t-statistica P Values Result
H1 0.28 2.97*** 0.003 Supported
H2 0.41 6.02*** 0.000 Supported
H3 0.35 4.05*** 0.000 Supported
H4 0.22 2.67*** 0.008 Supported
H5 -0.27 2.86*** 0.004 Supported
H6 -0.09 0.65 0.51 Not supported
H7 -0.007 0.08 0.93 Not supported
H8 0.34 2.17** 0.03 Supported
H9 0.44 3.88*** 0.000 Supported
According to the results of PLS modeling, all of the hypotheses were supported except
hypothesis 6 and 7. Privacy concerns antecedent was found to have a statistically neg-
ative significant impact on trust and positive significant impact on perceived risk. Trust
antecedent was also found to have a statistically significant positive impact on contin-
uance intention of Facebook users. In addition, effort expectancy and social influence
antecedents were found to have significant positive impacts on both trust and continu-
ance intention. As seen from Table 3, the highest path coefficient is 0.44 that belongs
to the path of trust-continuance intention. This means that if Facebook wants its users
to be loyal then it must build trust. The answer of “how to build trust” question is also
355
Business and Society
6
seen from the path coefficients. Facebook must focus on the dimensions of social in-
fluence, effort expectancy and privacy concerns in order to build trust, respectively.
3 Conclusion
This study helps draw useful implications for the managers of Facebook and social
networking services as well. It is important for practitioners to know the impacts of
antecedents on continuance intention to develop more efficient strategies. In this re-
search, both effort expectancy and social influence significantly affected trust anteced-
ent. According to the findings, privacy concerns was found to be one of the important
antecedents of continuance intention. Privacy concerns negatively influenced trust an-
tecedent. Besides, trust antecedent was found to have a statistically significant impact
on continuance intention of Facebook users. On the other hand, perceived risk was not
found to have a significant impact on both trust and continuance intention. According
to our findings, effort expectancy and social influence also affected continuance inten-
tion significantly. These results implicate that user experience in a social networking
service (in this case Facebook) is so crucial. Social networking services should focus
on designing positive and unique experiences for users and these experiences should be
shared by the users voluntarily. In addition, privacy concerns is very important for Fa-
cebook users so that managers of Facebook should be more aware of this finding and
develop strategies to be more confidential.
References
1. Sura, S., Ahn, J., & Lee, O.: Factors influencing intention to donate via social network site
(SNS): From Asian’s perspective. Telematics and Informatics 34(1), 164-176 (2017).
2. Chang, S. E., Liu, A. Y. & Shen, W. C.: User Trust in Social Networking Services: A Com-
parison of Facebook and LinkedIn. Computers in Human Behavior 69, 207-217 (2017).
3. Mouakket, S.: Factors influencing continuance intention to use social network sites: The
Facebook case. Computers in Human Behavior 53, 102-110 (2015).
4. Wang, Y., Min, Q., & Han, S.: Understanding the effects of trust and risk on individual
behavior toward social media platforms: A meta-analysis of the empirical evidence. Com-
puters in Human Behavior 56, 34-44 (2016).
5. Tan, X., Qin, L., Kim, Y., & Hsu, J.: Impact of privacy concern in social networking web
sites. Internet Research 22(2), 211-233 (2012).
6. Huang, H. Y., Chen, P. L., & Kuo, Y. C.: Understanding the facilitators and inhibitors of
individuals’ social network site usage. Online Information Review 41(1), 85-101 (2017).
7. Malik, A., Hiekkanen, K., Dhir, A., & Nieminen, M.: Impact of privacy, trust and user ac-
tivity on intentions to share Facebook photos. Journal of Information, Communication and
Ethics in Society 14(4), 364-382 (2016a).
8. Malik, A., Hiekkanen, K., & Nieminen, M.: Privacy and trust in Facebook photo sharing:
age and gender differences. Program 50(4), 462-480 (2016b).
356
Business and Society
7
9. Thelwall, M., & Vis, F.: Gender and image sharing on Facebook, Twitter, Instagram, Snap-
chat and WhatsApp in the UK: Hobbying alone or filtering for friends?. Aslib Journal of
Information Management 69(6), 702-720 (2017).
10. Padyab, A., & Ståhlbröst, A.: Exploring the dimensions of individual privacy concerns in
relation to the Internet of Things use situations. Digital Policy, Regulation and Governance
20(6), 528-544 (2018).
11. Hajli, N., Sims, J., Zadeh, A. H., & Richard, M. O.: A social commerce investigation of the
role of trust in a social networking site on purchase intentions. Journal of Business Research
71, 133-141 (2017).
12. F. Hair Jr, J., Sarstedt, M., Hopkins, L., & G. Kuppelwieser, V.: Partial least squares struc-
tural equation modeling (PLS-SEM) An emerging tool in business research. European Busi-
ness Review 26(2), 106-121 (2014).
13. Fornell, C., & Larcker, D. F.: Structural equation models with unobservable variables and
measurement error: Algebra and statistics. Journal of Marketing Research 18(3), 382-388
(1981).
14. Cortina, J. M.: What is coefficient alpha? An examination of theory and applications. Journal
of Applied Psychology 78, 98-104 (1993).
357