Developing a mobile SNS addiction scale utilizing factor analysis and the Rasch model
| Date | 11 November 2019 |
| Published date | 11 November 2019 |
| DOI | https://doi.org/10.1108/OIR-10-2018-0300 |
| Pages | 1284-1301 |
| Author | Xinghua Wang |
| Subject Matter | Library & information science,Information behaviour & retrieval,Collection building & management,Bibliometrics,Databases,Information & knowledge management,Information & communications technology,Internet,Records management & preservation,Document management |
Developing a mobile SNS
addiction scale utilizing factor
analysis and the Rasch model
Xinghua Wang
Normal College, Qingdao University, Qingdao, China
Abstract
Purpose –The purpose of this paper is to develop a mobile social networking service (SNS) addiction scale to
measure respondents’addiction levels.
Design/methodology/approach –Drawing on the existing literature on the components model of
addiction by Griffiths (2005) and mobile SNS addiction, an initial scale in a five-point Likert-format was
developed. It was refined through the pilot study with 100 participants and the main study with 423
participants utilizing factor analysis and Rasch analysis.
Findings –Mobile SNS addiction as a behavioral addiction, demonstrated six addiction symptoms:
modification, salience, tolerance, withdrawal, conflict and relapse, which were interrelated with eachother. The
mobileSNS addiction scaledeveloped in thisstudy was found to be psychometricallyrobust and unidimensional.
Practical implications –The mobile SNS addiction scale consists of nine items, thus making it easier and
more convenient to be applied to academic research and clinical practice.
Originality/value –The combined use of factor analysis and the Rasch model could largely reduce potential
negative effects associated with limitations of classical test theory and improve the chance of developing a
psychometrically robust instrument. The mobile SNS addiction scale covers a range of types of SNSs, thus
being more generic. The items in the scale are unidimensionally loaded on the latent construct of mobile SNS
addiction and demonstrate measurement invariance across respondents of different demographics.
Keywords Addiction, Factor analysis, Scale, Mobile SNSs, The Rasch model
Paper type Research paper
1. Introduction
Social networking services (SNSs) are virtual communities where users create individual
public profiles, communicate and interact with real-life friends, meet other people who share
similar intereststo them and keep abreast of whatis happening in their social circles( Ma etal.,
2018; Kuss and Griffiths, 2011; Mäntymäkiand Islam, 2016). There has beena proliferation of
studies on SNSs in recent years with the popularity of SNS providers worldwide such as
Facebook, Snapchat, WhatsApp and WeChat.Nevertheless, limited researchhas investigated
the dark sides of SNSs, in particular, addictions to SNSs (Yang et al., 2016).
SNS addiction is defined as a maladaptive psychological state of SNS use that is
represented in a repetitive usage pattern and a loss of control and is at the expense of other
important activities (LaRose et al., 2003; Turel and Serenko, 2012). It can cause a variety of
negative consequences in people’s work, studies, social relationships and cognitive
functioning, such as reduced self-regulation, decreased involvement in real-life communities
and worsened academic or work performance (e.g. Kuss and Griffiths, 2011; Hsiao, 2017;
Salehan and Negahban, 2013).
Nevertheless, there has been a paucity of studies specifically devoted to developing SNS
addiction scales despite the increasing attention on this issue in recent years (Andreassen,
2015). Extant scales measuring SNS addiction mostly focused on the use of Facebook (e.g.
Andreassen et al., 2012; da Veiga et al., 2019; Pontes et al., 2016). But the addiction to
Facebook is fundamentally different from that to SNSs (Griffiths, 2012), as Facebook has
diversified its functions so much that social networking is only one in them. Therefore,
Griffiths (2012) argued that the term “Facebook addiction”may have been obsolete as the
term “Internet addiction”because that people can be involved in a variety of activities
Online Information Review
Vol. 43 No. 7, 2019
pp. 1284-1301
© Emerald PublishingLimited
1468-4527
DOI 10.1108/OIR-10-2018-0300
Received 13 October 2018
Revised 1 March 2019
Accepted 22 July 2019
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/1468-4527.htm
1284
OIR
43,7
besides social networking in Facebook such as playing games, watching videos and online
shopping. Moreover, different SNSs cater to different groups of people. For instance,
Instagram and Snapchat attract younger people than Facebook (Larsson, 2018; Utz et al.,
2015). Therefore, Facebook addiction should not be identical with social networking
addiction (Andreassen et al., 2017; Griffiths, 2012).
In addition, extantscales for SNS addiction were developed solely relying onclassical test
theory (CTT), which is subject to such limitations as sample and test dependence (Higgins,
2007; Van Zile-Tamsen, 2017). Therefore, there is a need for new statistical methods to
overcome these limitations and a call fora new scale to measure addiction to SNSs in general
instead of a particular website like Facebook (Griffiths, 2012; Monacis et al., 2017).
In this regard, this study aimed to fill the research gap by developing a generic SNS
addiction scale with robust psychometric properties through new methods. Additionally,
due to the ubiquitous access to SNSs via mobile phones and tablets, the vast popularity of
personal mobile devices and the fast mobile internet, mobile SNSs are transforming the way
in which people interact and communicate with one another in both their professional and
leisure lives. Furthermore, the majority of social networking activities happen on mobile
devices (Lin and Lu, 2015), which thereby serve as a main route for people to be addicted to
SNSs. As such, this study specifically targeted mobile SNS addiction.
Drawing on the existing literature related to Griffiths’(2005) components model of
addiction and mobile SNS addiction, an initial scale was proposed, tested and refined using
the Rasch model and CTT such as exploratory factor analysis (EFA) and confirmatory
factor analysis (CFA). Specifically, two phases of studies were conducted based on 523
participants in total in China: the pilot study and the main study. The pilot study sought to
explore the factor structure of the initial scale through EFA, whilst the main study aimed to
confirm its factor structure and psychometric properties via EFA, CFA and Rasch analyses.
Following the two studies, a mobile SNS addiction scale with nine items was developed.
2. Theoretical framework
2.1 Extant scales measuring SNS addiction and a call for a generic one
Overall, there has been a limited number of scales developed specifically for SNS addiction
in extant studies (Andreassen, 2015; Griffiths, 2013), even though research on this issue has
expanded in recent years. Most of the scales have particularly targeted the use of Facebook,
for instance, Bergen Facebook Addiction Scale (BFAS; Andreassen et al., 2012), Portuguese
version of BFAS (Pontes et al., 2016) and Facebook Intrusion Questionnaire (Elphinston and
Noller, 2011). In addition, there are other scales modified from BFAS, such as Bergen Social
Media Addiction Scale (BSMAS; Andreassen et al., 2017) and the Italian version of BSMAS
(Monacis et al., 2017), wherein Facebook was simply replaced with the term “social media.”
Moreover, it seems that Facebook addiction has been considered synonymous with social
networking addiction in many studies (e.g. da Veiga et al., 2019; De Cock et al., 2014).
Nevertheless, addiction to a specific commercial website is different from that to social
networking, especially because Facebook has a variety of features besides social
networking and caters to users of different demographics from other SNSs such as Snapchat
and Instagram (Griffiths, 2012; Larsson, 2018).
Furthermore, all these scales were developed only using CTT such as factor analysis
and correlational analysis, which has been criticized for deficiencies in generating
psychometrically robust instruments due to such limitations as sample dependence and test
dependence ( for more information, see Higgins, 2007; Van Zile-Tamsen, 2017).
In view of the limitations of CTT, the Rasch model is recommended to be used for a
thorough psychometric evaluation in high-stakes situations (Petrillo et al., 2015), such as
developing instruments for clinical or diagnostic purposes (e.g. measuring SNS addiction).
The Rasch (1960) model was developed to help researchers create and refine the functioning
1285
Developing a
mobile SNS
addiction scale
To continue reading
Request your trial