Original Article SMJ

An Assessment of the Validity and Reliability of the

Social-Media Addiction Screening Scale (S-MASS)
Sirinda Chanpen, M.Sc.*, Chanvit Pornnoppadol, M.D.*, Asara Vasupanrajit, M.Sc.**, Quankamon Dejatiwongse
Na Ayudhya, M.D., Ed.M.*
*Department of Psychiatry, Faculty of Medicine Siriraj Hospital, Mahidol University, Bangkok 10700, Thailand, **Department of Psychiatry, Faculty
of Medicine, Chulalongkorn University, Bangkok 10330, Thailand.

ABSTRACT
Objective: The excessive use of social media can lead to addiction among many vulnerable individuals. Hence, the
utilization of a valid and reliable screening test to assess social media addiction is warranted.
Materials and Methods: The Social-Media Addiction Screening Scale (S-MASS) is a newly developed, self-report
screening scale containing 16 items that assess the three main components of behavioral addiction: giving priority,
impaired control, and negative consequences. The S-MASS reliability was measured using Cronbach’s alpha. An
exploratory factor analysis (EFA) and a confirmatory factor analysis (CFA) were employed to assess the S-MASS
factorial validity. A latent profile analysis (LPA) was also carried out to identify the classes of problematic social
media users.
Results: In all, 5,068 participants aged ≥ 13 years were recruited from five high schools and an online survey.
Cronbach’s alpha for the S-MASS was 0.90 (95% CI: 0.89–0.90), indicating excellent test reliability. The EFA and
CFA revealed a good factorial validity for the S-MASS. Based on the LPA, the participants were classed as “low-
risk” (n = 1,227; 24.2%), “moderate-risk” (n = 2,757; 54.4%), and “high-risk” (n = 1,084; 21.4%) problematic social
media users. The key differences between these classes were gender, age, necessity to use social media for work,
self-perception of addiction, and time spent on social media.
Conclusion: The S-MASS is a valid and reliable screening scale for social media addiction. The criterion validity
of the S-MASS should be evaluated once formal diagnostic criteria for social media addiction become available.

Keywords: Assessment; social media; addiction; screening; test (Siriraj Med J 2023; 75: 167-180)

INTRODUCTION Growing evidence suggests problematic social media

In recent years, there has been a dramatic increase use or social media addiction (SMA) can cause negative
in the use of social media, defined as forms of electronic effects to vulnerable individuals similarly to those found
communication through which users create online in other behavioral addictions.2-4 SMA is associated with
communities to share various types of contents. Using several mental health conditions, including attention deficit
social media has become an essential part of the daily hyperactivity disorder (ADHD), obsessive compulsive
routines of many people. On average, people spend 2 disorder (OCD), depression, and anxiety.5-7 Hence, early
hours 24 minutes daily on social networks, excluding SMA detection and intervention in people with SMA-
other internet usage.1 People are spending increasing especially those with comorbid psychiatric disorders-
amount of time gaining the benefits from the platforms would yield improved patient outcomes.
while being exposed to the risks they bring. To date, social media addiction has not yet been

Corresponding author: Chanvit Pornnoppadol

E-mail: [email protected]
Received 22 September 2022 Revised 5 December 2022 Accepted 13 December 2022 All material is licensed under terms of
ORCID ID:http://orcid.org/0000-0002-5181-9679 the Creative Commons Attribution 4.0
International (CC-BY-NC-ND 4.0)
https://doi.org/10.33192/smj.v75i3.261044 license unless otherwise stated.

https://he02.tci-thaijo.org/index.php/sirirajmedj/index Volume 75, No.3: 2023 Siriraj Medical Journal 167

Chanpen et al.

established as an official clinical diagnosis in DSM guidelines The aim of this study was to develop a new screening
or ICD-11. Neither its clinical definition has arrived at a scale for the assessment of SMA-namely, the Social-Media
consensus.2 The term is also used widely in non-clinical Addiction Screening Scale (S-MASS). The authors based
contexts.8 However, its operational definition in clinical the development of the S-MASS on the operational
study is often derived from Griffith’s six core components definition of behavioral addiction by ICD-11, for the
of behavioral addiction: salience, mood modification, purpose of clinical diagnosis relevancy. The authors set
tolerance, withdrawal symptoms, conflict, and relapse.2,5 out to comprehensively explore the reliability and validity
ICD-11 also describes three key features of disorders of the S-MASS, based on a large-scale and heterogeneous
due to addictive behaviors as 1) impaired control, a sample with a wide age range, and to find an appropriate
persistent pattern of repetitive behavior in which the cut-off point to identify high-risk problematic social
individual exhibits impaired control over the behavior; media users. Once the psychometric properties of this
2) increasing priority, given to the behavior to the extent new screening instrument are established, S-MASS will
that it takes precedence over other life interests and daily be another useful tool for epidemiological and clinical
activities; and 3) negative consequences, continuation, or studies of SMA.
escalation of the behavior despite negative consequences.9
This research operationalized social media addiction using MATERIALS AND METHODS
ICD-11 essential features of disorder due to addictive Participants
behaviors for the benefit of clinical relevancy. Participants were included if they were at least
The integration of factors is the likely explanation of 13 years old, which is the minimum age required to
social media addiction: dispositional, sociocultural and register for most social networking sites. Participants
reinforcing behavior factors.2 Neurological and personality needed to have used social media for at least 3 months
factors are example of dispositional explanation. From preceding the study. Participants were randomly recruited
neurological point of view, both chemical and behavioral from 2 sources: 1) five high schools in Bangkok; and
addiction is explained through brain’s reward systems.2,10 2) an online survey posted on the Facebook fan page
While personality factors often refer to the big-five of the Division of Child and Adolescent Psychiatry,
personalities and their correlations with social media Department of Psychiatry, Faculty of Medicine Siriraj
addiction.2,11 Sociocultural perspectives posit that certain Hospital, Mahidol University. Over a six-month study
family dynamics such as parental pressure influences period, 5,437 participants were recruited. 369 participants
SMA.2,12 However, more research is yet to be done. Lastly, were excluded due to incomplete S-MASS data.
SMA is also explained through some learning theories
such as operant conditioning, classical conditioning, Measures and Procedure
and social learning.2,12,13 For instance, positive outcomes The S-MASS is a newly developed, 16-item, self-report
from using social media, namely, entertainment and questionnaire to assess the severity of SMA. The initial
attention, are positive reinforcements that influence the item pool was generated by the principal investigator. The
same behavior (using social media) to be more likely to final items were selected and reduced by the consensus
repeat. of all investigators. The development of the S-MASS
Many previously validated SMA measurement tools was theoretically based on three key features of ICD-11
(except for the Bergen Social Media Addiction Scale) behavioral addictions. Items in each domain were derived
were developed using small, homogeneous, and narrow from the 9 criteria for Internet Gaming Disorder (IGD)
age-range samples, which mostly comprised adolescents that are outlined in the Diagnostic and Statistical Manual
or young adults.5,14-24,25-28,29-33 Moreover, approximately of Mental Disorders (DSM-5)34, including preoccupation,
half of the tools were specifically designed for Facebook withdrawal, tolerance, unsuccessful attempts to control,
addiction, rather than SMA generally, and many lack a loss of interests, continued excessive use, deception,
comprehensive factor-structure assessment (Table 1). escape, and jeopardized function. The jeopardized function
In addition, only a few used standard statistical analyses was extended beyond previously validated tools by also
to identify appropriate cut-off scores to differentiate asking about disturbed functions in areas of life other
problematic from normal social media use. Therefore, than relationships. An item having to do with deceitful
there is the need for a measurement tool that is applicable behavior was omitted from the final pool of items due
to social media use in general, which has been validated to it having the least sensitivity among all criteria.35
with a larger sample size and more-standardized analytical A content validation process was then performed.
methods. All 16 S-MASS items were examined for their relevance,

168 Volume 75, No.3: 2023 Siriraj Medical Journal https://he02.tci-thaijo.org/index.php/sirirajmedj/index

 Original Article SMJ
TABLE 1. Comparison among measurements developed to assess social media addiction.

Year Name of measurement Country No. of items No. of Participants Age, years Cronbach’s alpha
participants mean ± SD (range) coefficient

2010 Addictive Tendencies Scale (ATS)14 Australia 3 201 College students 19.1 ± 1.9 (17–24) 0.76

2011 Facebook Intrusion Australia 8 342 Undergraduate 19.8 ± 1.8 (18–25) 0.85
Questionnaire (FIQ)15 students

2012 Bergen Facebook Addiction Norway 6 423 College students 22.0 ± 4.0 (N/A) 0.83
Scale (BFAS) 16
18 (original)

2012 Social Networking Website USA 5 194 College students – 0.86

Addiction Scale (SNWAS) 17

2013 Facebook Dependence Peru 8 418 College students 20.1 ± 2.5 (N/A) 0.67
Questionnaire (FDQ) 18

2013 Facebook Addiction Scale (FAS)19 Turkey 8 447 College students 21.6 ± 1.9 (18–30) 0.86

2013 Addictive Tendencies Toward China 20 316 Adults 26.6 ± 4.4 (18–40) 0.92
Social Networking Sites 20

2015 Arabic Social Media Addiction Kuwait 14 1,327 Undergraduate 21.9 ± N/A (18–31) 0.61–0.75
Scale (SMAS) 21
students

2015 Bergen Facebook Addiction Scale Thailand 6 874 High school students 16.7 ± 1.0 (N/A) 0.91
– Thai Version (Thai-BFAS)22 (10th–12th grade)

2016 Facebook Addiction Test Germany and 7 (short) 1,019 Online survey 27.5 ± 9.1 (N/A) 0.92 (long)
(F-AT) 23
Austria 20 (long)

2016 Social Media Disorder (SMD) Netherlands 9 (short) 2,198 Online survey 14.05 ± 2.1 (10–17) 0.76 (short)
Scale 24
27 (long) 14.36 ± 2.2 (10–17) 0.90–0.92 (long)

2016 Bergen Social Media Addiction Norway 6 23,533 Online survey 35.8 ± 13.3 (16–88) 0.88
Scale (BSMAS)5 Hungary 6 5,961 High school students 16.62 ± 0.96 (15–22) 0.85
(9th–10th grade)

https://he02.tci-thaijo.org/index.php/sirirajmedj/index Volume 75, No.3: 2023 Siriraj Medical Journal 169

Chanpen et al.

TABLE 1. Comparison among measurements developed to assess social media addiction. (Continued)

Year Name of measurement Country No. of items No. of Participants Age, years Cronbach’s alpha
participants mean ± SD (range) coefficient

2017 Bergen Social Media Addiction Italy 6 734 High school and 21.6 ± 3.9 (16–40) 0.88
Scale – Italian version
25
college students

2017 Bergen Social Media Addiction Iran 6 2,676 High school 15.5 ± 1.2 (14–19) 0.86
Scale – Persian version 26
students

2018 Chinese Social Media Addiction China 58 619 College students 20.4 ± 1.5 (18–25) 0.94
Scale (Liu & Ma, 2018) 27
28

2018 Turkish Adaptation of the Social Turkey 9 553 Adolescents N/A (14–18) 0.83–0.86
Media Disorder Scale in Adolescents28

2018 Cross-Sectional and Longitudinal German 9 Study 1: 192 Phone interview Study 1 Female: 22 (21–24) 0.690–0.774
Evaluation of the Social Network Study 2: 2,316 and online Study 1 Male: 23 (21–27)
Use Disorder and Internet Gaming survey Adults Study 2 Female: 32 (25–27)
Disorder Criteria
29
Study 2 Male: 37 (27–53)

2019 Psychometric Testing of Three China 6 (BSMAS) 307 University students 21.64 ± 8.11 (17–30) 0.819
Chinese Online-Related Addictive
Behavior Instruments among
Hong Kong University Students30

2019 Cross-cultural validation of the China 9 903 College students 20.56 ± 2.75 (N/A) 0.753
Social Media Disorder scale31

2019 Spanish version of the Facebook Spain 8 567 Adults 29.09 ± 12.03 (18–67) 0.9
Intrusion Questionnaire (FIQ-S) 32

2020 Social Networking Addiction India 24 Study 1: 525 N/A 20.33 ± 1.70 (17-25) N/A
Scale (SNAS) 33
Study 3: 334

Abbreviations: N/A, not available; SD, standard deviation

170 Volume 75, No.3: 2023 Siriraj Medical Journal https://he02.tci-thaijo.org/index.php/sirirajmedj/index

 Original Article SMJ
clarity, and simplicity by a total of eight experts consisting (CFI). To indicate a good fit of the model, the CFI values
of two general psychiatrists, three child and adolescent had to be > 0.95, while the values for RMSEA and SRMR
psychiatrists, two clinical psychologists, and an expert were required to be < .06 and < .08, respectively.37
in social media communication. Each expert rated each Participants were classified into groups based on their
item using one-to-four Likert scale on three measures: S-MASS scores. The classification model was derived from
relevance, clarity, and simplicity. Rating score of three latent profile analysis (LPA) or a Gaussian finite mixture
or four by an expert was counted as one score, resulting model by expectation-maximization (EM) algorithm. The
in eight score points at most and zero point at least R package mclust (R Foundation, Vienna, Austria) was also
for each item. The score was converted into the item used in the classification process. Bayesian Information
content validity index or I-CVI for each item (Max = 1.00, Criterion (BIC), Bootstrap Likelihood Ratio Test (BLRT),
Min = 0.00). The content validity index for the whole scale Log- likelihood, and Integrated Complete-data Likelihood
was an average of all I-CVIs (Max = 1.00, Min = 0.00). (ICL) were conducted as means to accurately reveal number
Focus group interviews were conducted with a of latent classes. A better model fit is reflected in lower
group of ten high school students ages from fifteen to BIC and AIC values, and higher log likelihood values.
eighteen years old to ensure clarity of understanding The BLRT was conducted to compare models’ fitness.
by those completing the scale. Each student was given a Models with significant changes in −2 log-likelihood
printed copy of the scale. They were provided five minutes implied greater fitness to the data.
to go through the whole paper. Then, the investigators A Kruskal–Wallis test, followed by a Dunn–Bonferroni
asked how students interpreted each item, one by one. post-hoc pairwise comparison, were used to compare
The investigators took notes and used comments to the S-MASS scores among classes. Pearson’s chi-squared
adjust language to ensure each question item convey test was used to compare the categorical variables of
the intended meanings. the classes. Spearman’s rank correlation coefficient was
All items are scored on a four-point Likert scale also employed to measure the strengths of association
ranging from 0 (“definitely not true”) to 3 (“definitely between the average daily time spent on social media
true”). The total S-MASS score, obtained by summing and the S-MASS scores. All tests of significance were two
the participants’ responses, ranges from 0-48. The higher tailed, and a p-value < 0.05 was considered statistically
the S-MASS score is, the greater the likelihood that a significant.
participant is addicted to social media.
Ethics
Statistical analysis The study was approved by the Institutional Review
Data were analyzed using PASW Statistics for Board of the Faculty of Medicine Siriraj Hospital (COA
Windows (version 18; SPSS Inc., Chicago, Ill., USA). no. Si 701/2013). The Declaration of Helsinki was obliged
Cronbach’s alpha was used to measure the internal by investigators throughout the study. Informed consents
consistency reliability. A Cronbach’s alpha coefficient were obtained from all subjects prior to participating in
between 0.7 and 0.9 indicated acceptable reliability.36 the study. Parental consents were obtained in case of
Content validity was measured using a content validity minorities under 18 years old.
index (CVI) generated by an expert committee.
The sample of 5,068 subjects was randomly divided RESULTS
into two using the SPSS to produce two groups: one for Demographic statistics
an exploratory factor analysis (EFA; n = 2,534), and the Table 2 details the sociodemographic characteristics
other for a confirmatory factor analysis (CFA; n = 2,534). of the participants, the majority of whom were adolescents
The EFA was performed using SPSS. and young adults. Participants from each age group show
To test the three-factor model of the S-MASS, the different social media use patterns such as estimated
CFA was performed with LISREL for Windows (version hours of use per day (p<0.05). Younger participants
9.10; Scientific Software International Inc., Skokie, Ill., had higher S-MASS scores than older participants
USA). The model included three factors reflecting giving (p < 0.001). Two-thirds of participants (66.8%) reported
priority, impaired control, and negative consequences. The that it was necessary for them to use social media for
model’s goodness-of-fit was assessed with the following their work or study. Interestingly, approximately 30%
indices: chi-square test of model fit, standardized root of participants stated that they perceived themselves
means square residual (SRMR), root mean square error to be addicted to social media. The five most popular
of approximation (RMSEA), and comparative fit index social media platforms used were Facebook (87.9%),

https://he02.tci-thaijo.org/index.php/sirirajmedj/index Volume 75, No.3: 2023 Siriraj Medical Journal 171

Chanpen et al.

TABLE 2. Sociodemographic characteristics of participants and Social Media Addiction Screening Scale (S-MASS)
score.

Characteristics n (%) S-MASS score P-value

mean ± SD

Source of recruitment
Total 5068 (100.0) 22.58 ± 9.67
High schools 3672 (72.5) 23.17 ± 9.54 < 0.001
Online surveys 1396 (27.5) 21.05 ± 9.86

Gender
Male 2058 (41.0) 21.52 ± 9.30 < 0.001
Female 2963 (59.0) 23.28 ± 9.84

Age [mean ± SD (range) = 19.93 ± 9.96 (13–75) yrs]

13–17 yrs 3406 (69.8) 23.21 ± 9.50a < 0.001
18–25 yrs 525 (10.8) 22.94 ± 9.60a
26–45 yrs 760 (15.6) 20.78 ± 9.85b
46 yrs and above 186 (3.8) 16.26 ± 8.39c

Necessity to use social media for work

Necessary 3029 (66.8) 22.95 ± 9.73 < 0.001
Not necessary 1503 (33.2) 21.62 ± 9.59

Self-perception of addiction
Not addicted 1092 (22.8) 14.65 ± 7.49a < 0.001
Probably addicted 2245 (46.8) 21.90 ± 7.56 b

Addicted 1462 (30.5) 29.70 ± 8.68c

Amount of time spent on social media (not including gaming)

Weekdays [mean ± SD (range) = 3.23 ± 2.63 (0.08–18) hrs/day]
Light use (< 2 hrs/day) 1148 (26.7) 17.87 ± 8.65a < 0.001
Moderate use (2–4 hrs/day) 2264 (52.6) 22.92 ± 8.79 b

Heavy use (> 4 hrs/day) 892 (20.7) 27.80 ± 9.68c

Weekends [mean ± SD (range) = 4.91 ± 4.12 (0.08–24) hrs/day]
Light use (< 3 hrs/day) 1367 (31.7) 17.63 ± 8.66a < 0.001
Moderate use (3–5 hrs/day) 1692 (39.2) 22.68 ± 8.38 b

Heavy use (> 5 hrs/day) 1255 (29.1) 27.65 ± 9.30c

Average daily time spent [mean ± SD (range) = 3.68 ± 2.77 (0.08–19.71) hrs/day]
Light use (< 3 hrs/day) 2126 (50.6) 19.38 ± 8.66a < 0.001
Moderate use (3–5 hrs/day) 1203 (28.6) 24.06 ± 8.68 b

Heavy use (> 5 hrs/day) 876 (20.8) 28.31 ± 9.50c

Note: Different superscript letters (a, b, c) in the same column reflect a significant (p-value < 0.05) difference between the means, while the
same letter in one column reflects a non-significant difference between the means.

172 Volume 75, No.3: 2023 Siriraj Medical Journal https://he02.tci-thaijo.org/index.php/sirirajmedj/index

 Original Article SMJ
Line (79.2%), YouTube (77.0%), Google+ (38.7%), and A principal components analysis (PCA) revealed three
Instagram (37.6%). components that the authors named “giving priority”,
“negative consequences”, and “impaired control”; they
Correlation with time spent on social media had eigenvalues exceeding 1 and explained 39.53%, 7.38%,
A moderately positive correlation was observed and 6.41% of the variance, respectively (Table 3). The
between average daily time spent on social media and promax rotation method with Kaiser normalization was
S-MASS scores (rs38 = 0.412; p < 0.001). The heavy- chosen because the factor correlation matrix for all three
use group (using social media for > 4 hours/day on factors exceeded 0.32, which indicated that the factors
weekdays, or for > 5 hours/day on weekends) had higher in the analysis were correlated. All 16 items had factor
S-MASS scores than the light- and moderate-use groups loadings greater than 0.4. In addition, the Cronbach’s
(p < 0.001; Table 2). alphas for giving priority, negative consequences, and
impaired control were 0.789, 0.770, and 0.803, respectively,
Content validity indicating internal consistency of these three components
The calculated content validity index or CVI for the (Table 3).
S-MASS relative to the relevance, clarity, and simplicity
of all items were 0.992, 0.938, and 0.977, respectively. CFA
For items with item content validity index (I-CVI) less The three-factor model with the sixteen components
than 1.00, experts gave specific comments by identifying as indicator variables was tested with a CFA. The analysis
words which might be unclear to interpretation and provided an acceptable fit to the data (χ2 = 120.77, df =
suggested alternative words. Some experts also rewrote 62, p < 0.001; CFI = 0.99; RMSEA = 0.01; SRMR = 0.01).
or rearranged questions as examples. Others pointed out Factor loadings ranged from 0.32 to 0.74 (Table 4 and
double-barreled questions. Fig 1). Item 9 and item 13 of the S-MASS had factor
loadings less than 0.4 (0.37 and 0.32, respectively).
Reliability analysis
The S-MASS had excellent internal consistency, LPA
with a Cronbach’s alpha coefficient of 0.90 (95% CI: The LPA was performed on the sixteen items of the
0.89–0.90). In an item analysis, all 16 S-MASS items S-MASS, and according to the Bayesian Information
had a corrected item-total correlation above 0.3, which Criterion (BIC) and integrated complete-data likelihood
confirmed that each of the 16 items was correlated with (ICL) criterion, the three-class solution was selected as
the overall scale. If deleted, no item had a Cronbach’s the best-fitting model. The features of the three classes are
alpha greater than 0.90; this suggested that no item presented in Fig 2. The three classes were named “low-
disproportionately affected the overall reliability. risk” (S-MASS scores 0–15), “moderate-risk” (S-MASS
scores 16–30), and “high-risk” (S-MASS scores 31–48),
EFA representing 24.2% (n = 1,227), 54.4% (n = 2,757), and
Prior to performing the EFA, the suitability of the 21.4% (n = 1,084) of social media users, respectively.
data for the factor analysis was evaluated. Five factorability In the “high-risk” class, item 11— “I ignore or fail
assessment criteria were applied. First, the correlation matrix when people tell me to cut down my social media use”—
revealed the presence of several (82.03%) coefficients of showed elevated levels compared to the other items, as
0.30 or above, suggesting appropriate factorability. Second, shown in Fig 2. Interestingly, 67.6% of the participants
the Kaiser–Meyer–Olkin measure of sampling adequacy in the high-risk class believed that they were addicted
was 0.943, which exceeded the recommended value of to social media, whereas only 26.2% and 6.9% in the
0.6.39 Third, Bartlett’s test of sphericity achieved statistical moderate-risk and low-risk classes, respectively, believed
significance (χ240 = 29,879.585; p < 0.001), supporting so (Table 5).
the factorability of the correlation matrix. Fourth, the
diagonals of the anti-image correlation matrix were over DISCUSSION
0.5, suggesting the inclusion of each item in the factor The purpose of the present study was to measure
analysis. Finally, each item shares common variances the psychometric properties of the S-MASS, a newly
with other items demonstrated by the communalities developed, 16-item, self-report questionnaire developed
above 0.3 (Table 3). Given all of the above criteria were to screen for SMA. Results showed that the S-MASS
satisfied, all items of the S-MASS were included in the has excellent internal consistency (Cronbach’s alpha
EFA. coefficient, α = 0.90). This indicated that all 16 items

https://he02.tci-thaijo.org/index.php/sirirajmedj/index Volume 75, No.3: 2023 Siriraj Medical Journal 173

Chanpen et al.

TABLE 3. Exploratory factor analysis (EFA) of Social Media Addiction Screening Scale (S-MASS).

Loadings Communality
Questions Factor 1: Factor 2: Factor 3:
Giving priority Negative Impaired control
consequences
Since I started using social media…
(15) My friends regularly see me online. 0.730 0.572
(14) I feel that social media is a part of 0.686 0.564
my life that I can't lose.
(1) I use social media whenever I have a chance. 0.677 0.506
(7) I keep checking all the time to see if anyone 0.613 0.435
has “liked” or commented on the pictures/
statuses I have posted.
(2) I use social media as soon as I wake up 0.603 0.421
in the morning.
(13) I use social media to ease my stress. 0.596 0.393
(11) I ignore or fail when people tell me to 0.763 0.587
cut down my social media use.
(16) People around me say I am addicted 0.705 0.580
to social media.
(10) I use social media during circumstances 0.688 0.490
when I should not use it (e.g., while in the
classroom, doing daily activities, working,
meeting with friends or colleagues, walking
on the sidewalk, driving, etc.).
(12) I get agitated or irritable when I can’t 0.670 0.505
use social media.
(8) I talk to people on social media more 0.645 0.420
often than in real life.
(9) My social media use negatively impacts 0.626 0.563
my life in some ways.
(5) I often spend more time using social media 0.799 0.647
than I originally intended to.
(3) I spend all of my free time using social media. 0.791 0.653
(4) I have lost interest in other activities. 0.762 0.602
(6) I spend more time using social media now 0.732 0.593
than I used to.
Eigen value 6.324 1.181 1.026
Percentage of variance explained (total = 53.322) 39.528 7.384 6.410
Cronbach’s alpha 0.789 0.770 0.803

Extraction method: principal component analysis

Rotation method: Promax with Kaiser normalization
Factor loadings < 0.4 were suppressed

174 Volume 75, No.3: 2023 Siriraj Medical Journal https://he02.tci-thaijo.org/index.php/sirirajmedj/index

 Original Article SMJ

TABLE 4. Factor loadings, R2, and factor loading coefficient of Social Media Addiction Screening Scale (S-MASS).

Components Factor loadings R2 Factor loading

b B SE (b) t coefficient
Giving priority
(1) I use social media whenever I have 0.53 0.61 0.02 31.44* 0.37 0.17
a chance. (PRE1)
(2) I use social media as soon as I wake 0.62 0.57 0.02 28.93* 0.32 0.08
up in the morning. (PRE2)
(7) I keep checking all the time to see if 0.70 0.67 0.02 34.56* 0.45 0.20
anyone has “liked” or commented on the
pictures/statuses I have posted. (PRE3)
(13) I use social media to ease my stress. 0.32 0.36 0.02 17.13* 0.13 0.03
(PRE4)
(14) I feel that social media is a part of 0.70 0.72 0.02 37.78* 0.52 0.25
my life that I can't lose. (PRE5)
(15) My friends regularly see me online. 0.63 0.65 0.02 34.51* 0.42 0.15
(PRE6)

Negative consequences
(8) I talk to people on social media more 0.58 0.63 0.02 31.62* 0.40 0.22
often than in real life. (IMP1)
(9) My social media use negatively impacts 0.37 0.40 0.02 17.94* 0.15 0.04
my life in some ways. (IMP2)
(10) I use social media during circumstances 0.51 0.53 0.02 26.00* 0.28 0.08
when I should not use it (e.g., while in the
classroom, doing daily activities, working,
meeting with friends or colleagues, walking
on the sidewalk, driving, etc.). (IMP3)
(11) I ignore or fail when people tell me to 0.63 0.68 0.02 33.31* 0.46 0.17
cut down my social media use. (IMP4)
(12) I get agitated or irritable when I can’t 0.65 0.65 0.02 33.17* 0.42 0.14
use social media. (IMP5)
(16) People around me say I am addicted 0.70 0.71 0.02 37.86* 0.50 0.17
to social media. (IMP6)

Impaired control
(3) I spend all of my free time using social 0.73 0.75 0.02 40.51* 0.57 0.31
media. (LOC1)
(4) I have lost interest in other activities. 0.65 0.71 0.02 37.58* 0.50 0.25
(LOC2)
(5) I often spend more time using social 0.66 0.69 0.02 37.23* 0.48 0.18
media than I originally intended to. (LOC3)
(6) I spend more time using social media 0.74 0.79 0.02 40.82* 0.63 0.38
now than I used to. (LOC4)

*p-value < 0.01

https://he02.tci-thaijo.org/index.php/sirirajmedj/index Volume 75, No.3: 2023 Siriraj Medical Journal 175

Chanpen et al.

Fig 1. Factor loadings of Social-Media Addiction Screening Scale (S-MASS)

Fig 2. The three classes of social-media users obtained from the latent profile analysis

176 Volume 75, No.3: 2023 Siriraj Medical Journal https://he02.tci-thaijo.org/index.php/sirirajmedj/index

 Original Article SMJ
TABLE 5. Comparison of the three latent classes.

Characteristics Total Low-risk Moderate-risk High-risk Overall test

(n = 5,068) (n = 1,227) (n = 2,757) (n = 1,084) χ2(df) P-value

Gender (Female); n (%) 2,963 (59.0) 687 (56.4)a 1557 (57.0)a 719 (67.2)b 37.77 (2) < 0.001

Age (years) Median 16.00, min 13.00, 19.93 (9.96) 22.34 (11.91)a 19.49 (9.63)b 18.25 (7.55)c 34.30 (2)* < 0.001

max 75.00, SE 0.14: Mean (SD)

Necessity to use social media for work (Necessary); n (%) 3,029 (66.8) 715 (63.8)a 1,621 (66.3)a 693 (71.7)b 15.51 (2) < 0.001

Self-perception of addiction n (%) < 0.001

Probably addicted 2,245 (46.8) 452 (39.2)a 1,493 (57.0)b 300 (29.2)c 475.48 (2) < 0.001

Addicted 1,462 (30.5) 80 (6.9)a 687 (26.2)b 695 (67.6)c 1,042.90 (2) < 0.001

Amount of time spent on social media

(not including gaming) < 0.001

Weekdays (hrs/day) Median 2.50, min 0.08, 3.23 (2.63) 2.24 (1.97) a
3.16 (2.42) b
4.57 (3.20) c
526.81 (2)* < 0.001
max 18.00, SE 0.04: Mean (SD)

Weekends (hrs/day) Median 4.00, min 0.08, 4.91 (4.12) 3.05 (2.79)a 4.78 (3.60)b 7.41 (5.27)c 672.17 (2)* < 0.001

max 24.00, SE 0.06: Mean (SD)

Average daily time spent (hrs/day) Median 2.86, 3.68 (2.77) 2.46 (1.97)a 3.61 (2.51)b 5.28 (3.37)c 614.49 (2)* < 0.001

min 0.08, max 19.71, SE 0.04: Mean (SD)

Total Social Media Addiction Screening Scale 22.58 (9.67) 10.48 (3.72)a 22.51 (4.10)b 36.47 (4.52)c 4,133.80 (2)* < 0.001

(S-MASS) Score

Median 22.00, min 0, max 48, SE 0.14: Mean (SD)

Note: Different superscript letters (a, b, c) in the same row reflect a significant (p-value < 0.05) difference between the means, while same superscript letters in one row reflect a non-significant difference
between the means, according to the Pearson chi-square or (*) Kruskal–Wallis test, followed by a Tukey post-hoc pairwise comparison.

https://he02.tci-thaijo.org/index.php/sirirajmedj/index Volume 75, No.3: 2023 Siriraj Medical Journal 177

Chanpen et al.

of the S-MASS measured the same problem, namely, from which the participants were recruited, especially the
social media addiction (SMA). The EFA using a PCA online survey, allowed us to recruit and enroll participants
with promax rotation demonstrated that the S-MASS with relatively diverse sociodemographic backgrounds
has good factorial validity as all items had loading factors and a wider age range (13–75 years) than other studies
above 0.4. The EFA also revealed that 3 factors were (although the majority of participants were aged between
foundational to the S-MASS and covered the essential 13 and 25 years; Table 1). The authors set out to develop
characteristics of behavioral addiction (giving priority, the S-MASS for use in screening for SMA in general,
impaired control, and negative consequences). The CFA not Facebook addiction only. S-MASS can, therefore, be
also confirmed the three-factor model of the S-MASS. applied to subgroups of social media users who interact
According to the LPA, three classes of social media with social networking sites other than Facebook.
users were identified, based on their risk of addiction: high
risk, moderate risk, and low risk. Members of the high-risk Limitations
class were likely to (i) be female, (ii) be younger, (iii) have The present study has some limitations. First, results
necessary work-related use, (iv) perceive themselves as were based on a convenient sample, limiting the extent to
being addicted to social media, and (v) spend more than which findings can be generalized to a broader population.
5 hours daily on social media (Table 5). These risk factors Second, the S-MASS is a self-report questionnaire which
will help clinicians accurately identify social media users is subject to several biases (such as social desirability and
who might be at risk for SMA. Interestingly, item 11 of short-term recall). Third, the test-retest reliability was
the S-MASS (“I ignore or fail when people tell me to cut not evaluated to determine stability. Forth, the same cut-
down my social media use”) showed an elevated level in off might not be a one-size-fit-all since each age group
the high-risk class relative to the other items. This item shows different patterns of social media usage. Further
may be helpful in distinguishing the high-risk class from studies to identify age-specific cut-offs are warranted.
the low-risk and medium-risk classes. Furthermore, 21.4% Finally, although a few forms of validity were tested in
(n = 1,084) of participants belonged to the high-risk class, this study, other important types of validity should also
based on the LPA; this is consistent with the prevalence be examined (e.g., concurrent, predictive, convergent,
of social-networking-site addiction (29.5%) reported by and discriminant validities).
Tang and Koh.41 On the other hand, the high-risk class
proportion found in our study is somewhat lower than Future studies
the previously reported prevalence of Facebook addiction The S-MASS should be further validated—most
in Thailand (41.8%)42 and much higher than the rate notably, its criterion validity. This validation process
for the at-risk group of SMA in Hungary (4.5%).43 The can be undertaken after a formal diagnosis of SMA
disparity in the prevalence of SMA among countries is becomes available. In its present form, the S-MASS is
probably due to differences in the measurement tools best suited for use in epidemiologic studies. However,
and the sample populations used by the various studies. to test whether the S-MASS is sensitive to change after
Nevertheless, cultural influence might also contribute interventions is an interesting and worthwhile research
to the disparity found between countries.44 pursuit. Once the S-MASS is proven to be adequately
The study also discovered that the S-MASS score sensitive to change, it can also be used in clinical or
is moderately positively correlated with average daily interventional studies.
time spent on social media (rs38 = 0.412; p < 0.001). In
other words, the greater the amount of time spent on CONCLUSION
social media, the greater the risk of becoming addicted The Social-Media Addiction Screening Scale (S-MASS)
to social media. This finding may imply that the S-MASS is a psychometrically reliable and valid screening test for
can determine the severity of addiction. SMA. Two cut-offs are identified for risk classification.
Strengths of this study include its relatively large Further studies assessing the concurrent, predictive,
sample size, the heterogeneity of the participants, and convergent, and discriminant validity of the S-MASS
the comprehensive assessment of S-MASS reliability and in a more heterogeneous population are warranted. In
validity. Cut-off scores were also identified for three-level addition, the criterion validity of the S-MASS to determine
risk classification. This fulfills the gap in previous studies its sensitivity, specificity, and the appropriateness of the
on assessment tools. More importantly, SMA manifests current recommended cut-off scores should be evaluated
in a spectrum, not in binary categories. Having cut-points once formal diagnostic criteria for SMA become available.
is beneficial for clinical practice. The diversity of sources

178 Volume 75, No.3: 2023 Siriraj Medical Journal https://he02.tci-thaijo.org/index.php/sirirajmedj/index

 Original Article SMJ
ACKNOWLEDGEMENTS Cyberpsychol Behav Soc Netw. 2011;14(11):631-5.
The authors gratefully acknowledge Ms.Naratip 16. Andreassen C, Torsheim T, Brunborg G, Pallesen S. Utvikling
av en Facebook-avhengighetsskala. Psychol Rep. 2012;110(2):
Sanguanpanich, Ms.Thapanee Sangsawang, and Dr.
501-17.
Sasima Tongsai for their assistance with the statistical 17. Turel O, Serenko A. The benefits and dangers of enjoyment
analyses. with social networking websites. European Journal of Information
Systems. 2012;21(5):512-28.
Conflict of interest: The authors declare no conflict of 18. Wolniczak I, Caceres-DelAguila JA, Palma-Ardiles G, Arroyo
interest. KJ, Solís-Visscher R, Paredes-Yauri S, et al. Association between
Facebook dependence and poor sleep quality: a study in a sample
of undergraduate students in Peru. PLoS One. 2013;8(3):e59087.
REFERENCES 19. Koc M, Gulyagci S. Facebook addiction among Turkish college
1. Georgiev D. How Much Time Do People Spend on Social students: the role of psychological health, demographic, and
Media? 2022 [Available from: https://review42.com/how- usage characteristics. Cyberpsychol Behav Soc Netw. 2013;16(4):
much-time-do-people-spend-on-social-media/. 279-84.
2. Andreassen CS. Online social network site addiction: A 20. Wu AM, Cheung VI, Ku L, Hung EP. Psychological risk factors
comprehensive review. Current Addiction Reports. 2015;2(2): of addiction to social networking sites among Chinese smartphone
175-84. users. J Behav Addict. 2013;2(3):160-6.
3. Griffiths MD, Kuss DJ, Demetrovics Z. Social networking 21. Al-Menayes J. Psychometric properties and validation of the
addiction: An overview of preliminary findings. In: Kenneth Arabic social media addiction scale. J Addict. 2015;2015:291743.
PR, Laura CF, editors. Behavioral Addictions: criteria, evidence, 22. Phanasathit M, Manwong M, Hanprathet N, Khumsri J, Yingyeun
and treatment: Academic Press; 2014. p. 119-41. R. Validation of the Thai version of Bergen Facebook addiction
4. Hou Y, Xiong D, Jiang T, Song L, Wang Q. Social media scale (Thai-BFAS). J Med Assoc Thai. 2015;98:108-17.
addiction: Its impact, mediation, and intervention. Cyberpsychology: 23. Dantlgraber M, Wetzel E, Schützenberger P, Stieger S, Reips
Journal of Psychosocial Research on Cyberspace. 2019;13(1). U-D. Simple construct evaluation with latent class analysis: An
5. Andreassen CS, Billieux J, Griffiths MD, Kuss DJ, Demetrovics investigation of Facebook addiction and the development of
Z, Mazzoni E, et al. The relationship between addictive use of a short form of the Facebook Addiction Test (F-AT). Behav
social media and video games and symptoms of psychiatric Res Methods. 2016;48(3):869-79.
disorders: A large-scale cross-sectional study. Psychol Addict 24. van den Eijnden RJ, Lemmens JS, Valkenburg PM. The Social
Behav. 2016;30(2):252-62. Media Disorder Scale. Computers in Human Behavior. 2016;
6. Robinson A, Bonnette A, Howard K, Ceballos N, Dailey S, 61:478-87.
Lu Y, et al. Social comparisons, social media addiction, and 25. Monacis L, De Palo V, Griffiths MD, Sinatra M. Social networking
social interaction: An examination of specific social media addiction, attachment style, and validation of the Italian
behaviors related to major depressive disorder in a millennial version of the Bergen Social Media Addiction Scale. J Behav
population. Journal of Applied Biobehavioral Research. Addict. 2017;6(2):178-86.
2019;24(1):e12158. 26. Chung-Ying Lin AB, Per Nilsen, Mark D Griffiths, Amir H
7. Shensa A, Sidani JE, Dew MA, Escobar-Viera CG, Primack BA. Pakpour. Psychometric validation of the Persian Bergen Social
Social media use and depression and anxiety symptoms: A Media Addiction Scale using classic test theory and Rasch
cluster analysis. Am J Health Behav. 2018;42(2):116-28. models. J Behav Addict. 2017;6(4):620-9.
8. Sun Y, Zhang Y. A review of theories and models applied in 27. Liu C, Ma J. Development and validation of the Chinese
studies of social media addiction and implications for future social media addiction scale. Personality and Individual
research. Addict Behav. 2021;114:106699. Differences. 2018;134:55-9.
9. World Health Organization. ICD-11 for mortality and morbidity 28. Savci M, Ercengiz M, Aysan F. Turkish adaptation of the
statistics 2018 [Available from: https://icd.who.int/browse11/l-m/ Social Media Disorder Scale in adolescents. Noro Psikiyatr
en#/http://id.who.int/icd/entity/1448597234. Ars. 2018;55(3):248-55.
10. Karim R, Chaudhri P. Behavioral Addictions: An Overview. 29. Bouna-Pyrrou P, Aufleger B, Braun S, Gattnar M, Kallmayer S,
J Psychoactive Drugs. 2012;44(1):5-17. Wagner H, et al. Cross-sectional and longitudinal evaluation
11. Kircaburun K, Griffiths MD. Instagram addiction and the Big of the social network use disorder and internet gaming disorder
Five of personality: The mediating role of self-liking. J Behav criteria. Front Psychiatry. 2018;9:692.
Addict. 2018;7(1):158-70. 30. Yam C-W, Pakpour AH, Griffiths MD, Yau W-Y, Lo C-LM,
12. Smith MA. Social learning and addiction. Behav Brain Res. Ng JM, et al. Psychometric testing of three Chinese online-
2021;398:112954. related addictive behavior instruments among Hong Kong
13. Wang X. Mobile SNS Addiction as A Learned Behavior: A university students. Psychiatr Q. 2019;90(1):117-28.
Perspective from Learning Theory. Media Psychology. 31. Fung SF. Cross-cultural validation of the Social Media Disorder
2020;23(4):461-92. scale. Psychol Res Behav Manag. 2019;12:683-90.
14. Wilson K, Fornasier S, White KM. Psychological predictors 32. Bendayan R, Blanca Mena MJ. Spanish version of the Facebook
of young adults' use of social networking sites. Cyberpsychol intrusion questionnaire (FIQ-S). Psicothema. 2019;31(2):204-9.
Behav Soc Netw. 2010;13(2):173-7. 33. Shahnawaz MG, Rehman U. Social Networking Addiction
15. Elphinston RA, Noller P. Time to face it! Facebook intrusion and Scale. Cogent Psychology. 2020;7(1):1832032.
the implications for romantic jealousy and relationship satisfaction. 34. American Psychiatric Association. Diagnostic and statistical

https://he02.tci-thaijo.org/index.php/sirirajmedj/index Volume 75, No.3: 2023 Siriraj Medical Journal 179

Chanpen et al.

manual of mental disorders (DSM-5®): American Psychiatric Journal of Contemporary Psychotherapy. 2009;39(4):241-6.
Association; 2013. 41. Tang CS-K, Koh YYW. Online social networking addiction
35. Ko CH, Yen JY, Chen SH, Wang PW, Chen CS, Yen CF. among college students in Singapore: Comorbidity with behavioral
Evaluation of the diagnostic criteria of Internet gaming disorder addiction and affective disorder. Asian J Psychiatr. 2017;25:
in the DSM-5 among young adults in Taiwan. J Psychiatr Res. 175-8.
2014;53:103-10. 42. Khumsri J, Yingyeun R, Mereerat M, Hanprathet N, Phanasathit
36. Tavakol M, Dennick R. Making sense of Cronbach’s alpha. M. Prevalence of Facebook Addiction and Related Factors
Int J Med Educ. 2011;2:53-5. Among Thai High School Students. J Med Assoc Thai. 2015;98:
37. Brown TA. Confirmatory factor analysis for applied research: S51-60.
Guilford publications; 2015. 43. Bányai F, Zsila Á, Király O, Maraz A, Elekes Z, Griffiths MD, et al.
38. Griffiths M. A ‘components’ model of addiction within a Problematic Social Media Use: Results from a Large-Scale Nationally
biopsychosocial framework. Journal of Substance Use. 2005; Representative Adolescent Sample. PloS One. 2017;12(1):e0169839.
10(4):191-7. 44. Cheng C, Lau Y-C, Chan L, Luk JW. Prevalence of social
39. Kaiser HF, Rice J. Little Jiffy, Mark IV. Educational and media addiction across 32 nations: Meta-analysis with subgroup
psychological measurement. Educational and Psychological analysis of classification schemes and cultural values. Addict
Measurement 1974;34(1):111-7. Behav. 2021;117:106845.
40. Young K. Internet addiction: diagnosis and treatment considerations.

180 Volume 75, No.3: 2023 Siriraj Medical Journal https://he02.tci-thaijo.org/index.php/sirirajmedj/index