International Journal of

Environmental Research
and Public Health

Article
Problematic Internet Usage and Self-Esteem in Chinese
Undergraduate Students: The Mediation Effects of Individual
Affect and Relationship Satisfaction
Guang Zeng 1 , Lijin Zhang 1 , Sai-fu Fung 2, * , Jingwen Li 3 , Yi-Man Liu 4,5 , Zi-Ke Xiong 5 , Zhi-Quan Jiang 5 ,
Fang-Fang Zhu 6 , Zhen-Ting Chen 7 , Si-Ding Luo 8,9 , Ping Yu 8,9 and Qian Huang 10

1 Department of Psychology, Sun Yat-sen University, Guangdong 510275, China;

zengg6@[Link] (G.Z.); zhanglj37@[Link] (L.Z.)
2 Department of Social and Behavioural Sciences, City University of Hong Kong, Hong Kong, China
3 Department of Psychology, Tsinghua University, Beijing 100084, China; lijw921012@[Link]
4 School of Economics, Finance and Marketing, Royal Melbourne Institute of Technology,
Melbourne, VIC 3001, Australia; [Link]@[Link]
5 Guangzhou Huashang College, University Administrative Office, Guangdong 511300, China;
xzk_scg_cn@[Link] (Z.-K.X.); jzq_edu_cn@[Link] (Z.-Q.J.)
6 School of Economics and Trade, Guangzhou Huashang College, Guangdong 511300, China; tjx_jr@[Link]
7 School of Data Science, Guangzhou Huashang College, Guangdong 511300, China; z.t_chen@[Link]
8 Managing Director Office, Global Business College of Australia, Melbourne, VIC 3000, Australia;



[Link]@[Link] (S.-D.L.); [Link]@[Link] (P.Y.)

 9 General Manager Office, Edvantage Institute Australia, Melbourne, VIC 3000, Australia
10 Department of Sports Training, Xi’an Physical Education University, Xi’an 710068, China;
Citation: Zeng, G.; Zhang, L.; Fung,
huangqian168@[Link]
S.-f.; Li, J.; Liu, Y.-M.; Xiong, Z.-K.;
* Correspondence: sffung@[Link]
Jiang, Z.-Q.; Zhu, F.-F.; Chen, Z.-T.;
Luo, S.-D.; et al. Problematic Internet
Abstract: The aim of this cross-sectional study was to examine the mediating effects of individual
Usage and Self-Esteem in Chinese
affect and relationship satisfaction on the relationship between self-esteem and Problematic Internet
Undergraduate Students: The
Mediation Effects of Individual Affect
Use (PIU). Affect was measured using the Positive and Negative Affect Schedule (PANAS), relation-
and Relationship Satisfaction. Int. J. ship satisfaction was assessed using a positive and negative semantic dimension scale, self-esteem
Environ. Res. Public Health 2021, 18, was measured using the Rosenberg Self-Esteem Scale, and PIU was measured using the Problematic
6949. [Link] Internet Use scale with a sample of 507 Chinese university students (Mage = 20.41 years, SD = 2.49).
ijerph18136949 The relationships between the variables were tested using structural equation modelling with a
multiple mediation model. The results revealed that negative affect and the negative semantic
Academic Editor: Gergely Fehér dimensions of relationship satisfaction mediated the relationship between self-esteem and PIU. The
implications of the results and the study’s theoretical contributions are discussed.
Received: 1 June 2021
Accepted: 25 June 2021
Keywords: cognitive–behavioural model; Internet addiction; self-esteem; multiple mediation model;
Published: 29 June 2021
positive and negative affect schedule

Publisher’s Note: MDPI stays neutral

with regard to jurisdictional claims in
published maps and institutional affil-
1. Introduction
iations.
The Internet was first introduced to the public as a global computer network in
the early 1990s [1–3]. The Internet is indispensable for communication, information
exchange and sharing, entertainment, socialising, teaching and learning and business
development [4–6], due to advances in technology and the ubiquity of broadband and
Copyright: © 2021 by the authors.
Licensee MDPI, Basel, Switzerland.
mobile phones. However, despite its importance and ease of use, the Internet has the
This article is an open access article
potential to cause harm, as widely acknowledged in academia [7–13]. Its many advantages
distributed under the terms and
may be cancelled out by potential addictive behaviour or problematic use [14].
conditions of the Creative Commons Problematic Internet Use (PIU) is defined as excessive online activity likely to be
Attribution (CC BY) license (https:// associated with significant functional impairment, including compulsive online shopping,
[Link]/licenses/by/ gambling, cybersex, and excessive online streaming and social media use that are addictive,
4.0/). impulsive or compulsive elements [15]. Research has shown that problematic internet use

Int. J. Environ. Res. Public Health 2021, 18, 6949. [Link] [Link]
Int. J. Environ. Res. Public Health 2021, 18, 6949 2 of 14

is associated with the neglect of important life areas, such as declining educational and
work achievement, decreasing sleeping time, reduced quality of meals, and a narrowing
range of interests [16–18]. An excessive amount of Internet use also has a negative effect on
family and partner relations and on communication within the family [19]. Problematic
internet use can also cause risk of physical health of individuals [20] or form the basis for
criminal activity [21]. Impulsive and compulsive traits can underlie problematic internet
behavior [15,22–24], during specific internet activities have been associated with psychiatric
disorders. For example, online shopping has been linked to depression and hoarding [24].
Grant, et al. [25] noted that addictive behaviour can be conceptualised as substance or
non-substance addiction. Problematic Internet Use (PIU) is considered to be a type of non-
substance behavioural addiction. Demetrovics, et al. [26] developed a PIU questionnaire,
which has become a well-established scale for evaluating addictive Internet behaviour
worldwide, including in Africa [27], the Americas [28], Asia [29,30] and Europe [31,32].
The questionnaire was designed to identify the extent of this problem, including whether
it is uncontrollable, whether it is markedly distressing and whether it results in social
difficulties or hypomania [33]. Factor analysis of the questionnaire of PIU identifies three
components: obsession (mental engagement with the Internet), neglect (neglect of everyday
activities and essentials needs) and control disorder (difficulties in controlling Internet
use) [26]. Other measures have been developed to evaluate specific Internet-based addictive
behaviour among adolescents and young adults, including gaming [34,35], gambling [36],
social media [37], smartphone use [38] and cybersex [39,40].
Demetrovics, Szeredi and Rozsa [26] developed a PIU questionnaire, which has be-
come a well-established scale for evaluating addictive Internet behaviour worldwide, in-
cluding in Africa [27], the Americas [28], Asia [29,30] and Europe [31,32]. The questionnaire
was designed to identify the extent of this problem, including whether it is uncontrollable,
whether it is markedly distressing and whether it results in social difficulties or hypoma-
nia [33]. Identifying the mechanisms of PIU can be challenging. The problem has most
commonly been approached from the cognitive–behavioural perspective [41,42]. A cogni-
tive behavioural model of internet addiction suggests that low self-esteem is a core aspect
of maladaptive cognition of the self, which is the critical factor for problematic internet
use [41]. Caplan [42] noted that the ‘virtual interpersonal world might present a serious risk
to some individuals’ (p. 573). Other studies have focused on the failure of individuals to
control their Internet use and its consequences for social relationships [43–45] and academic
performance [46,47]. Senol-Durak and Durak [48] claimed that psychosocial well-being,
particularly self-esteem, can reduce the likelihood of PIU. Importantly, research has linked
self-esteem and PIU in such that low self-esteem is a risk factor for PIU [49] and self-esteem
often has identified as a precursor to PIU [50–53]. A Bangladeshi survey study found that
self-esteem is correlated with problematic internet use [52]. There are several studies that
also suggest that problematic internet use exerts negative impacts on self-esteem among
India adolescents [53], Spanish adolescents [50], and Korean adolescents [51].
Although many studies have found direct relationships between self-esteem and
PIU, more research about the internal mechanism is required [54]. Exploring mediating
factors might shed light on the relationship between self-esteem and PIU, and advance our
understanding of the aetiology of PIU. In this study, we answered the call made by Li, Liu
and Dong [54] for further investigation of the internal mechanisms related to self-esteem
and PIU. Moreover, we extended the work of Davis [41], Caplan [42] and Senol-Durak
and Durak [48] by empirically demonstrating with a sample of university students that
individual affect and relationship satisfaction can be significant mediators of PIU.

1.1. Self-Esteem and Problematic Internet Use

‘Self-esteem’ refers to one’s subjective evaluation of one’s own worth, and has been
extensively examined by psychologists, criminologists and sociologists [55–58]. According
to the risk-buffering hypothesis, favourable individual traits, such as self-esteem, can
attenuate the relationship between risk factors and psychological status of individual [59].
Int. J. Environ. Res. Public Health 2021, 18, 6949 3 of 14

Individuals with high levels of self-esteem demonstrated better psychological adjustment

such as low negative affect and high positive affect [60], better academic performance [61],
lower level of depression and anxiety [62], improved health behaviour and relationship
satisfaction [63]. Rosenberg, Schooler and Schoenbach [61], who pioneered the concept of
self-esteem and first operationalised its measurement, illustrated its positive effects on aca-
demic performance and problems in adolescence. Pyszczynski, Solomon, Greenberg, Arndt
and Schimel [62] later suggested that high levels of self-esteem can reduce anxiety and
related defensive behaviour, along with recurrent thoughts about death. Further positive
impacts of self-esteem have since been identified, such as enhancing social support among
in-patients [36], serving as a preventative measure against anxiety during the COVID-19
pandemic [64] and mitigating psychological distress experienced by infertile couples [65].
Promoting self-esteem can help to create positive workplace environments [66] and reduce
behaviour detrimental to health [67]. In a cross-cultural study, Diener and Diener [63]
identified positive associations between self-esteem and satisfaction with friends, family
and life relationships.
Research suggests that self-esteem serves as a preventative factor of PIU [68]. A high
level of self-esteem has been found to alleviate the adverse effects of PIU [69–73]. A recent
study found that social adjustment and self-esteem had indirect effects, through procras-
tination, on Internet addiction among college students [74]. Additionally, research has
revealed that self-esteem may be indirectly related to academic performance, mental health
and problematic social media usage [75,76], and that a combination of self-esteem, a sense
of meaning in life and reflection can help reduce the impact of Internet-related addictive
behaviour [77,78]. Peng, et al. [79] further suggested that self-esteem was closely associ-
ated with Internet addiction, and self-esteem has a mediation effect on the relationship
between feeling disconnected from school and Internet addiction, based on a sample of
2758 Chinese adolescents. Therefore, it is important to further investigate the relationship
between self-esteem and PIU as well as its underlying mechanism.

1.2. Relationship Satisfaction, Individual Affect and Problematic Internet Use

The cognitive–behavioural perspective has been applied in many empirical studies
to explore the relationship between relationship satisfaction and PIU. Research found
that relationship satisfaction can mediate the effects of PIU [69,80–83]. Kim and Jun [81]
found that the link between Internet use and depression was mediated by relationship
satisfaction in a longitudinal study of 4576 older adults. Park, Kang and Kim [69] suggested
that children’s relationships with their parents and peers can influence the formers’ PIU.
Jeong, Kim, Yum and Hwang [83] argued that PIU can be exacerbated by poor parent–child
relationships and low self-esteem. A study of 2758 Chinese adolescents found that school
disconnectedness, which refers to students negative connections with their peers, teachers,
and community, can exacerbate Internet addiction [79].
Individual affect can also influence Internet-related behaviour. A relationship be-
tween negative affect and PIU has also been reported by several empirical studies [84–89].
Matthews, Farnsworth and Griffiths [88] revealed that negative mood states can predict
students’ involvement in online gambling, a specific category of PIU. Negative mood states
were experienced as a consequence of gambling, which in turn facilitated the need for
many of the gamblers to modify their mood through play again [90]. Guadagno, Rempala,
Murphy and Okdie [89] examined emotional responses to Internet videos and investi-
gated the mechanisms underlying these responses using the Positive and Negative Affect
Schedule (PANAS). They found that ‘only content that generates stronger affective re-
sponses are likely to spread as a viral video’ (p. 2318). Thus, examining the relationships
among self-esteem, individual emotional affect and relationship satisfaction can enhance
understanding of the mechanisms of PIU.
Int. J. Environ. Res. Public Health 2021, 18, 6949 4 of 14

1.3. Gender Differences of Problematic Internet Use

Gender seems to be an important correlate of PIU, and it is one of the most frequently
studies variables in the PIU literature [91]. In the most recent meta-analytical study, it
examined 115 independent samples from 34 countries, reporting a global gender differ-
ence where men were found to be more prone to PIU symptoms or severity compared to
women [92]. Men are about five times more likely to have PIU than women [93]. A study
examined the relationship between internet use and gender in 2160 Chinese adolescents
and found that males have higher PIU symptoms [94]. Another study also found that men
in a sample of 2059 Chinese adolescents scored significantly higher than women in PIU
severity [95]. Similarly, males outperformed females in PIU severity in various cultures,
including Indian [96], Vietnamese [97], British [98], Romanian [99] and Bengali [100]. In
terms of specific categories of internet behaviours, study found that males more often
engaged in online role-play games, file downloading, shopping, gambling, visiting porno-
graphic websites, and indiscriminate surfing, whereas the females are more likely to engage
in information searching, chatting, e-mailing, and social networking [101].
While the research mentioned above demonstrated such results, many other studies
found no significant gender difference in PIU symptoms. A study investigates gender factor
in relationship among impulsivity, behavioural inhibition and PIU, and found no gender
difference in the severity of the PIU [54]. Similarly, there are many studies that found no
gender difference in the severity of PIU in Japanese [102], the United Arab Emirates [103],
Portuguese [104] samples. Given the inconsistent findings in the literature, it is important
to further investigate the gender factor on PIU.

1.4. Purpose of the Study

The purpose of this study was to explore the relationships among self-esteem, indi-
vidual affect, relationship satisfaction and PIU, and to test the mediating roles of both
individual affect and relationship satisfaction on the link between self-esteem and PIU.
Based on the results of previous studies, we proposed and tested the following three
hypotheses regarding the relationships among self-esteem, individual affect and relation-
ship satisfaction and their influence on PIU. We also tested whether the mediation effects
differed by gender through the multigroup mediation model.

Hypothesis 1 (H1). Self-esteem would be negatively related to problematic internet use.

Hypothesis 2 (H2). Negative affect would mediate the effect of self-esteem on problematic internet use.

Hypothesis 3 (H3). Relationship dissatisfaction would mediate the relationships between self-
esteem and problematic internet use.

Hypothesis 4 (H4). There is a gender difference on relationships among self-esteem, individual

affect, relationship satisfaction, and PIU.

2. Methods
2.1. Participants
We conducted our cross-sectional research in April and May 2019 as part of a broader
study related to quality of life and Internet use among students at a university in Guangzhou,
China. We recruited 507 undergraduate students (Mage = 20.41 years, SD = 2.49, 14.5%
male) through the university’s intranet system using a self-report smartphone-based appli-
cation. The recruited sample reflected the demographic profile of the university population.
The participants were asked to provide their informed consent before voluntarily partici-
pating in the study. The research process was conducted in accordance with the relevant
rules and regulations set out in the Statistics Law of the People’s Republic of China and the
Declaration of Helsinki, acknowledging their revisions.
Int. J. Environ. Res. Public Health 2021, 18, 6949 5 of 14

2.2. Measures
The PIU questionnaire [26] measured the three latent factors of obsession, neglect and
control disorder, with six items per subscale. The participants were asked to rate each item
on a 5-point Likert-type scale ranging from 1 = never to 5 = always. The questionnaire
has been validated in a Chinese context [30] and has been frequently used to analyse
PIU among Chinese adolescents [105]. The scale in this study had good reliability, with a
Cronbach’s α of 0.862.
PANAS [106] includes both positive and negative affect factors, with 10 items per
factor. Items are measured on a 5-point Likert-type scale. The overall reliability of PANAS
is 0.846. In this study, the Cronbach’s α values for the subscales of positive affect (PA) and
negative affect (NA) were 0.802 and 0.884, respectively, in line with those in other PANAS
studies in the context of Chinese society [65,107,108].
The positive and negative semantic dimensions of relationship satisfaction were
measured using a relationship satisfaction scale (PN-SMD) [109]. The participants were
asked to rate the 14 items (7 positive and 7 negative) on an 8-point Likert-type scale from
0 (not at all) to 7 (always). The PN-SMD has been used to examine deviant Internet
behaviour among Chinese adolescents [110] and behaviour manifesting alcohol addiction
in couples [111]. The Cronbach’s α values for the PN-SMD scale as a whole, the positive
semantic dimension (P-SMD) subscale and the negative semantic dimension (N-SMD)
subscale were 0.764, 0.905 and 0.922, respectively.
Self-esteem was measured using the 10-item Rosenberg Self-Esteem Scale (RSES) [61].
This is a 5-point Likert-type scale in which 0 = strongly disagree and 4 = strongly agree. The
Chinese version of the RSES has been validated [112–114] and the scale had good reliability
in the present study, with Cronbach’s α = 0.763.

2.3. Procedure
We first conducted confirmatory factor analysis (CFA) to ensure the validity of the
five scales. We treated the response data as continuous for all of the scales and conducted
our CFA with a maximum likelihood estimator, as the data were normally distributed, i.e.,
skewness < 2 and kurtosis < 7 [115,116].
The latent variable models were assessed using multiple fit indices. Our criteria were
that (1) the Tucker–Lewis index (TLI) and the comparative fit index (CFI) should both be
equal to or greater than 0.9 [117–119]; and (2) the root mean square error of approximation
(RMSEA) and standardized root mean square residuals (SRMR) should both be no more
than 0.08 [120].
The results of the validity testing are given in Table 1. The good model fit indicated
that the scales we used had high construct validity.

Table 1. Confirmatory factor analysis.

Scales CFI TLI SRMR RMSEA 90%CI of RMSEA

PIUQ 0.928 0.902 0.043 0.054 (0.046, 0.062)
PANAS 0.916 0.903 0.048 0.058 (0.052, 0.065)
PN-SMD 0.918 0.902 0.042 0.100 (0.091, 0.109)
RSE 0.942 0.913 0.053 0.066 (0.051, 0.081)

The correlations between the variables were consistent with our assumptions (Table 2).
The preliminary correlation results also supported the significant relationships observed
between the variables for the male and female participants and the magnitudes of these
relationships (Table 3). We then further analysed the mediation effects among the variables
through our multigroup mediation model.
Int. J. Environ. Res. Public Health 2021, 18, 6949 6 of 14

Table 2. Correlation matrix of all variables (n = 507).

Mean SD 1 2 3 4 5
1. PIUQ 47.945 9.828
2. Positive_Affect 29.892 5.833 −0.003
3. Negative_Affect 24.022 7.267 0.471 ** 0.191 **
4. Positive_SMD 32.616 7.655 −0.241 ** 0.377 ** −0.241 **
5. Negative_SMD 20.448 8.376 0.369 ** −0.131 ** 0.451 ** −0.502 **
6. RSE 26.667 3.935 −0.296 ** 0.275 ** −0.454 ** 0.420 ** −0.443 **
Note: * p < 0.05, ** p < 0.01.

Table 3. Correlation matrix of all variables (split by gender).

Male (n = 73) Female (n =434)

1 2 3 4 5 1 2 3 4 5
1. PIUQ
2. Positive_Affect 0.015 −0.007
3. Negative_Affect 0.526 ** 0.205 0.465 ** 0.187 **
4. Positive_SMD −0.352 ** 0.380 ** −0.300 ** −0.215 ** 0.377 ** −0.226 **
5. Negative_SMD 0.605 ** −0.016 0.566 ** −0.405 ** 0.307 ** −0.156 ** 0.424 ** −0.525 **
6. RSE −0.218 0.159 −0.418 ** 0.277 * −0.300 ** −0.321 ** 0.298 ** −0.459 ** 0.449 ** −0.475 **
Note: * p < 0.05, ** p < 0.01.

3. Results
3.1. Mediation Model
We developed a multiple mediation model [121] of the PIU questionnaire to simulta-
neously test the indirect effects of the PANAS and PN-SMD measures while controlling
for gender. The data analyses were conducted using Mplus 8.4 [122]. The model fit was
good (CFI = 0.930, TLI = 0.917, SRMR = 0.075, RMSEA = 0.064, 90% CI of RMSEA = [0.058,
0.070]). Table 4 and Figure 1 give the standardised coefficients of the multiple mediation
model, indicating that the relationship between RSE and PIU was mediated by negative
affect (β = −0.256, p < 0.001,95% CI = (−0.460, −0.221)) and negative SMD (β = −0.100,
p = 0.004, 95% CI = (−0.235, −0.029)).

Table 4. Standardised results of mediation effects.

Path β p 95% CI 1
RSE-Positive Affect-PIUQ −0.017 0.396 (−0.095, 0.034)
RSE-Negative Affect-PIUQ −0.256 0.000 (−0.460, −0.222)
RSE-Positive SMD-PIUQ −0.038 0.258 (−0.139, 0.051)
RSE-Negative SMD-PIUQ −0.100 0.004 (−0.235, −0.029)
195% confidence interval of bootstrapping procedure with 1000 bootstrap draws. Note: bold = significant
mediation effect.

Before implementing the multigroup mediation model, we tested the measurement

invariance of the model’s four scales between the male and female participants. First, we
tested the configural invariance of all of the scales and found that the structures of the four
scales were the same for the male and female samples, demonstrating a good model fit
(Tables 5–8). We then constrained the loadings to be the same for both genders (metric
invariance) and assessed whether the model’s fit was significantly different from the fit of
the configural invariance model (∆CFI should be less than 0.01) [123]. Our results showed
that all of the scales satisfied the requirement of metric invariance. We then applied scalar
invariance models in which the intercepts of the observed variables were set as equal for
the different groups. The results showed that the fit of the scalar invariance models was as
good as that of the metric invariance models (∆CFI < 0.01). Thus, measurement invariance
between the male and female participants was demonstrated for all of the models.
 for gender. The data analyses were conducted using Mplus 8.4 [122]. The model fit was
good (CFI = 0.930, TLI = 0.917, SRMR = 0.075, RMSEA = 0.064, 90% CI of RMSEA = [0.058,
0.070]). Table 4 and Figure 1 give the standardised coefficients of the multiple mediation
model, indicating that the relationship between RSE and PIU was mediated by negative
affect
Int. J. Environ. Res. Public Health 2021, (β = −0.256, p < 0.001,95% CI = (−0.460, −0.221)) and negative SMD (β = −0.100,7 pof =14
18, 6949
0.004, 95% CI = (−0.235, −0.029)).

Figure 1. Multiple mediation model. Note: * p < 0.05, ** p < 0.01. The solid lines represent significant
Figure 1. Multiple mediation model. Note: * p < 0.05, ** p < 0.01. The solid lines represent significant
path coefficients, and the bold solid lines indicate the mediation effects. The dotted lines represent
path coefficients, and the bold solid lines indicate the mediation effects. The dotted lines represent
non-significantpath
non‐significant pathcoefficients.
coefficients.

Table 5. Measurement invariance of RSE.

Model CFI TLI SRMR RMSEA 90%CI of RMSEA ∆CFI Pass?

Configural Invariance 0.938 0.903 0.055 0.070 (0.054, 0.086) Yes
Metric Invariance 0.940 0.919 0.059 0.064 (0.049, 0.079) 0.002 Yes
Scalar Invariance 0.941 0.930 0.061 0.060 (0.045, 0.075) 0.001 Yes

Table 6. Measurement Invariance of PIUQ.

Model CFI TLI SRMR RMSEA 90%CI of RMSEA ∆CFI Pass?

Configural Invariance 1.000 1.000 0.000 0.000 (0.000, 0.000) Yes
Metric Invariance 1.000 1.000 0.014 0.000 (0.000, 0.069) 0.000 Yes
Scalar Invariance 0.995 0.992 0.038 0.061 (0.000, 0.125) 0.005 Yes
Note: Due to the complexity of the model structure of PIUQ scale and the small sample size of the male group, the multigroup model did
not converge. Therefore, we packaged the 18 observed variables of PIUQ scale and then conducted multigroup model for this scale.

Table 7. Measurement invariance of PANAS.

Model CFI TLI SRMR RMSEA 90%CI of RMSEA ∆CFI Pass?

Configural Invariance 0.907 0.891 0.058 0.063 (0.056, 0.070) Yes
Metric Invariance 0.903 0.893 0.065 0.062 (0.055, 0.069) 0.004 Yes
Scalar Invariance 0.899 0.894 0.065 0.062 (0.055, 0.068) 0.004 Yes

Table 8. Measurement Invariance of PN-SMD.

Model CFI TLI SRMR RMSEA 90%CI of RMSEA ∆CFI Pass?

Configural Invariance 0.925 0.910 0.045 0.096 (0.087, 0.106) Yes
Metric Invariance 0.925 0.917 0.048 0.093 (0.084, 0.102) 0.000 Yes
Scalar Invariance 0.920 0.920 0.048 0.090 (0.082, 0.099) 0.005 Yes

3.2. Multigroup Mediation Model

We tested whether the mediation effects differed by gender through the multigroup
mediation model. Table 9 gives the mediation effects for male and female groups. We
used the Wald test to identify any differences in the mediation effects between the two
groups [123]. The mediation effect of negative SMD was found to differ significantly
(Wdif4 = −0.249, p = 0.039). For the male participants, it mediated the relationship between
Int. J. Environ. Res. Public Health 2021, 18, 6949 8 of 14

RSE and PIUQ (β4male = −0.252, p = 0.004, 95% CI = [−0.808, 0.059]), but this effect was not
significant for the female participants (β4female = −0.034, p = 0.396, 95% CI = [−0.130, 0.052]).
We found no significant differences in the other mediation effects (Wdif1 = 0.010, p = 0.781;
Wdif2 = 0.079, p = 0.521; W_dif3 = 0.004, p = 0.956). The mediation effect of negative SMD
was significant in the female sample (β2female = −0.255, p < 0.001, 95% CI = [−0.345, −0.174])
but not in the male sample (β2male = −0.167, p = 0.097, 95% CI = (−0.838, 0.086)). However,
this difference was not significant.

Table 9. Standardised results of mediation effects (split by gender).

Male Female
Path β p 95% CI 1 β p 95% CI 1
1. RSE-Positive Affect-PIUQ −0.007 0.769 (−0.170, 0.073) −0.018 0.460 (−0.070, 0.037)
2. RSE-Negative Affect-PIUQ −0.167 0.097 (−0.838, 0.086) −0.255 0.000 (−0.345, −0.174)
3. RSE-Positive SMD-PIUQ −0.032 0.553 (−0.286, 0.071) −0.039 0.331 (−0.123, 0.048)
4. RSE-Negative SMD-PIUQ −0.252 0.004 (−0.808, −0.059) −0.034 0.396 (−0.130, 0.052)
1 95% confidence interval of bootstrapping procedure with 1000 bootstrap draws. Note: bold = significant mediation effect.

4. Discussion
We examined the relationships among self-esteem, affect, relationship satisfaction
and PIU. Echoing previous studies, the findings of our bivariate and multivariate analyses
suggested that self-esteem was significantly negatively related to problematic internet
use [63,86]. Hypothesis 1 was therefore supported. The results reported in Table 2 found
a non-significant relationship between PA and PIU, with r = −0.003, whereas P-SMD
showed a significant negative relationship (r = −0.241, p < 0.01) with PIU. The observed
relationships and their magnitudes remained similar after the data were divided by gender
(Table 3). The multiple mediation model further confirmed these relationships (Figure 1
and Table 4). Neither positive affect nor relationship satisfaction was significantly related
to PIU, again supporting previous findings [82,124].
Our main contribution lies in our illustration of the mediating roles of both negative
individual affect and relationship dissatisfaction in the association between self-esteem
and PIU. Using a multiple mediation model, we found that negative scores for affect
(β = −0.256, p < 0.001) and relationship dissatisfaction (β = −0.100, p = 0.004) mediated
the relationship between self-esteem and PIU (Figure 1 and Table 4), supporting Hy-
pothesis 2 and 3. However, although both positive affect and relationship satisfaction
were related to PIU, neither mediated the relationship between self-esteem and PIU. Our
results also suggest a gender difference in terms of PIU (Table 9), which supports Hy-
pothesis 4. Our findings therefore provide empirical evidence in line with the cognitive–
behavioural perspective [41,42,125]. In short, the current study provides empirical evi-
dence to support the claim that ‘social isolation and lack of self-esteem might play in the
PIU process’ [42] (p. 569). As such, the findings further clarified the internal mechanisms
related self-esteem and PIU which reported in the existing literature [54], by illustrating
that negative affect and poor relationships mediate PIU.
Our study has several limitations. First, the sample was restricted to young adults,
but psychological mechanisms and behavioural patterns related to Internet use may vary
due to demographic and age differences. The cross-age stability and applicability of our
conclusions could be examined in future research using samples from various age groups.
Second, our sample is taken from a school in southern China. Due to China’s vast territory
and large population, there are also differences in population distribution between the
north and the south. There may be doubts whether our sampling can represent all college
students in China. Moreover, the ratio of males to females in the sample of this study
does not necessarily represent the national college students. Future research can take
research samples from all provinces in China, which can improve the generalizability of
research conclusions. Third, our main approaches were a self-report measure and a cross-
Int. J. Environ. Res. Public Health 2021, 18, 6949 9 of 14

sectional design, preventing us from establishing causality among the variables. In future
research, a longitudinal design could be used, with lab experiments conducted to examine
causality. Fourth, our proposed multiple mediation model indicated that negative affect
and relationship dissatisfaction only partially accounted for the effect of self-esteem on
PIU. Therefore, other factors may be involved in this effect. Other psychological variables
should therefore be explored in future studies.
Despite these limitations, we confirmed that negative affect and relationship dissatis-
faction mediated the relationship between self-esteem and PIU. This finding has significant
practical implications for mitigating PIU among young adults. Drawing on the cognitive–
behavioural model [41,42], our results indicated that negative emotions and interpersonal
dissatisfaction are closely related to PIU. Thus, practical and effective interventions and
training could help young adults alleviate their PIU, by enabling them to regulate their
emotions and build satisfactory interpersonal relationships. We also found interpersonal
dissatisfaction to be an important potential cause of PIU. Schools, teachers and parents
should therefore be aware that helping young adults develop satisfactory interpersonal
relationships is extremely important. Relevant interventions could include the teaching of
skills for and practical methods of developing such relationships. Education departments,
schools and parents should offer advice on and training in emotional management, main-
taining interpersonal relationships and ensuring healthy communication. They should also
organise activities to enhance young adults’ social relations and interpersonal satisfaction.
If young adults have a healthy emotional outlook, are able to regulate their emotions, and
have good interpersonal communication skills and sound interpersonal relationships, their
PIU is likely to diminish.

5. Conclusions
We investigated the relationships among self-esteem, individual affect, relationship
satisfaction, and PIU. Mediation analysis indicated that negative affect and relationship
dissatisfaction mediated the relationship between self-esteem and PIU. Interestingly, neither
positive affect nor relationship satisfaction mediated the relationship between self-esteem
and PIU. The present study identifies a mechanistic pathway (i.e., via negative affect,
relationship dissatisfaction) through which we can better understand the critical process
by which self-esteem buffers PIU. Therefore, PIU intervention programs in young adults
should focus more on relieving relationship dissatisfaction and negative affect than on
improving positive affect and relationship satisfaction.

Author Contributions: Conceptualization, G.Z., L.Z., S.-F.F., J.L.; Methodology, G.Z., L.Z., S.-F.F., J.L.,
Y.-M.L.; Validation, Y.-M.L., Z.-K.X., Z.-Q.J., F.-F.Z., Z.-T.C., S.-D.L., P.Y., Q.H.; Funding Acquisition,
G.Z., L.Z., S.-F.F., J.L., Y.-M.L.; Project Administration, S.-F.F., Y.-M.L., Z.-K.X., Z.-Q.J., F.-F.Z., Z.-T.C.,
S.-D.L., P.Y., Q.H.; Data Curation, S.-F.F., Y.-M.L., Z.-K.X., Z.-Q.J.; Formal Analysis, G.Z., L.Z., S.-F.F.,
J.L.; Investigation, G.Z., L.Z., S.-F.F., J.L.; Writing, G.Z., L.Z., S.-F.F., J.L. All authors have read and
agreed to the published version of the manuscript.
Funding: This research was supported by China Postdoctoral Science Foundation, grant number
2020M672920, China positive psychology research foundation, grant number 0020344, Sun Yat-Sen
University young teacher training basic scientific research fund, grant number 26000-31610133. The
funding body had no influence on the content of the paper and the publication process.
Institutional Review Board Statement: The research has gone through proper research ethical
approval procedures specified by Guangzhou Huashang College. Its procedure was in compliance
with the Declaration of Helsinki guidelines.
Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.
Data Availability Statement: The datasets used and analyzed during the current study are available
from the correspondent author on reasonable request.
Conflicts of Interest: The authors declare no conflict of interest.
Int. J. Environ. Res. Public Health 2021, 18, 6949 10 of 14

References
1. Payne, W.B.; O’Sullivanl, D. Exploding the Phone Book: Spatial Data Arbitrage in the 1990s Internet Boom. Ann. Am. Assoc. Geogr.
2020, 110, 391–398. [CrossRef]
2. Mullins, M.E.; Kozlowski, S.W.J.; SchMitt, N.; Howell, A.W. The role of the idea champion in innovation: The case of the Internet
in the mid-1990s. Comput. Hum. Behav. 2008, 24, 451–467. [CrossRef]
3. Murtishaw, S.; Schipper, L. Disaggregated analysis of US energy consumption in the 1990s: Evidence of the effects of the Internet
and rapid economic growth. Energy Policy 2001, 29, 1335–1356. [CrossRef]
4. Vayre, E.; Vonthron, A.M. Identifying Work-Related Internet’s Uses-at Work and Outside Usual Workplaces and Hours-and Their
Relationships with Work-Home Interface, Work Engagement, and Problematic Internet Behavior. Front. Psychol. 2019, 10, 15.
[CrossRef] [PubMed]
5. Li, S.M.; Chung, T.M. Internet function and Internet addictive behavior. Comput. Hum. Behav. 2006, 22, 1067–1071. [CrossRef]
6. Weiser, E.B. The functions of Internet use and their social and psychological consequences. CyberPsychol. Behav. 2001, 4, 723–743.
[CrossRef] [PubMed]
7. Pauwels, K.; Weiss, A. Moving from free to fee: How online firms market to change their business model successfully. J. Mark.
2008, 72, 14–31. [CrossRef]
8. Lewis, S.P.; Seko, Y. A Double-Edged Sword: A Review of Benefits and Risks of Online Nonsuicidal Self-Injury Activities. J. Clin.
Psychol. 2016, 72, 249–262. [CrossRef]
9. Tam, J.; Tang, W.S.; Fernando, D.J.S. The internet and suicide: A double-edged tool. Eur. J. Intern. Med. 2007, 18, 453–455.
[CrossRef]
10. Chan, T.H. Facebook and its Effects on Users’ Empathic Social Skills and Life Satisfaction: A Double-Edged Sword Effect.
Cyberpsychology Behav. Soc. Netw. 2014, 17, 276–280. [CrossRef] [PubMed]
11. Richards, J. What has the internet ever done for employees? A review, map and research agenda. Empl. Relat. 2012, 34, 22–43.
[CrossRef]
12. Sun, D.L.; Ma, N.; Bao, M.; Chen, X.C.; Zhang, D.R. Computer Games: A Double-Edged Sword? CyberPsychol. Behav. 2008, 11,
545–548. [CrossRef]
13. Matthews, M.; Murnane, E.; Snyder, J.; Guha, S.; Chang, P.; Doherty, G.; Gay, G. The double-edged sword: A mixed methods
study of the interplay between bipolar disorder and technology use. Comput. Hum. Behav. 2017, 75, 288–300. [CrossRef]
14. Rutkowski, L.; Svetina, D. Assessing the Hypothesis of Measurement Invariance in the Context of Large-Scale International
Surveys. Educ. Psychol. Meas. 2013, 74, 31–57. [CrossRef]
15. Ioannidis, K.; Treder, M.S.; Chamberlain, S.R.; Kiraly, F.; Redden, S.A.; Stein, D.J.; Lochner, C.; Grant, J.E. Problematic internet use
as an age-related multifaceted problem: Evidence from a two-site survey. Addict. Behav. 2018, 81, 157–166. [CrossRef]
16. Chou, C.; Condron, L.; Belland, J.C. A review of the research on Internet addiction. Educ. Psychol. Rev. 2005, 17, 363–388.
[CrossRef]
17. Nalwa, K.; Anand, A.P. Internet addiction in students: A cause of concern. CyberPsychol. Behav. 2003, 6, 653–656. [CrossRef]
18. Young, K.S. Caught in the Net: How to Recognize the Signs of Internet Addiction—And a Winning Strategy for Recovery; John Wiley
Sons: New York, NY, USA, 1998.
19. Kraut, R.; Patterson, M.; Lundmark, V.; Kiesler, S.; Mukopadhyay, T.; Scherlis, W. Internet paradox—A social technology that
reduces social involvement and psychological well-being? Am. Psychol. 1998, 53, 1017–1031. [CrossRef]
20. Király, O.; Griffiths, M.D.; Demetrovics, Z. Internet gaming disorder and the DSM-5: Conceptualization, debates, and controver-
sies. Curr. Addict. Rep. 2015, 2, 254–262. [CrossRef]
21. Recupero, P.R. Forensic Evaluation of Problematic Internet Use. J. Am. Acad. Psychiatry Law 2008, 36, 505–514.
22. Block, J.J. Issues for DSM-V: Internet addiction. Am. J. Psychiatry 2008, 165, 306–307. [CrossRef]
23. Cao, F.; Su, L.; Liu, T.; Gao, X. The relationship between impulsivity and Internet addiction in a sample of Chinese adolescents.
Eur. Psychiatry 2007, 22, 466–471. [CrossRef]
24. Claes, L.; Mueller, A.; Luyckx, K. Compulsive buying and hoarding as identity substitutes: The role of materialistic value
endorsement and depression. Compr. Psychiat. 2016, 68, 65–71. [CrossRef]
25. Grant, J.E.; Potenza, M.N.; Weinstein, A.; Gorelick, D.A. Introduction to Behavioral Addictions. Am. J. Drug Alcohol Abuse 2010,
36, 233–241. [CrossRef]
26. Demetrovics, Z.; Szeredi, B.; Rozsa, S. The three-factor model of Internet addiction: The development of the Problematic Internet
Use Questionnaire. Behav. Res Methods 2008, 40, 563–574. [CrossRef] [PubMed]
27. Opakunle, T.; Aloba, O.; Opakunle, O.; Eegunranti, B. Problematic Internet Use Questionnaire-Short Form-6 (PIUQ-SF-6):
Dimensionality, validity, reliability, measurement invariance and mean differences across genders and age categories among
Nigerian adolescents. Int. J. Ment. Health 2020, 49, 229–246. [CrossRef]
28. Kelley, K.J.; Gruber, E.M. Psychometric properties of the Problematic Internet Use Questionnaire. Comput. Hum. Behav. 2010, 26,
1838–1845. [CrossRef]
29. Lin, M.M.; Kim, Y. The reliability and validity of the 18-item long form and two short forms of the Problematic Internet Use
Questionnaire in three Japanese samples. Addict. Behav. 2020, 101, 8. [CrossRef]
Int. J. Environ. Res. Public Health 2021, 18, 6949 11 of 14

30. Koronczai, B.; Kökönyei, G.; Urbán, R.; Király, O.; Nagygyörgy, K.; Felvinczi, K.; Griffiths, M.D.; Huang, Z.; Demetrovics, Z.
Confirmation of the Chinese Version of the Problematic Internet Use Questionnaire Short Form (PIUQ-SF). Int. J. Ment. Health
Addict. 2017, 15, 191–197. [CrossRef]
31. Laconi, S.; Urban, R.; Kaliszewska-Czeremska, K.; Kuss, D.J.; Gnisci, A.; Sergi, I.; Barke, A.; Jeromin, F.; Groth, J.; Gamez-Guadix,
M.; et al. Psychometric Evaluation of the Nine-Item Problematic Internet Use Questionnaire (PIUQ-9) in Nine European Samples
of Internet Users. Front. Psychiatry 2019, 10, 13. [CrossRef]
32. Kern, L.; Acier, D. French adaptation of the Problematic Internet Use Questionnaire. Evol. Psychiatr. 2013, 78, 357–371. [CrossRef]
33. Shapira, N.A.; Goldsmith, T.D.; Keck, P.E.; Khosla, U.M.; McElroy, S.L. Psychiatric features of individuals with problematic
internet use. J. Affect. Disord. 2000, 57, 267–272. [CrossRef]
34. Cho, H.; Kwon, M.; Choi, J.H.; Lee, S.K.; Choi, J.S.; Choi, S.W.; Kim, D.J. Development of the Internet addiction scale based on the
Internet Gaming Disorder criteria suggested in DSM-5. Addict. Behav. 2014, 39, 1361–1366. [CrossRef] [PubMed]
35. Demetrovics, Z.; Urban, R.; Nagygyorgy, K.; Farkas, J.; Griffiths, M.D.; Papay, O.; Kokonyei, G.; Felvinczi, K.; Olah, A. The
Development of the Problematic Online Gaming Questionnaire (POGQ). PLoS ONE 2012, 7, e36417. [CrossRef]
36. Gonzalez-Cabrera, J.; Machimbarrena, J.M.; Beranuy, M.; Perez-Rodriguez, P.; Fernandez-Gonzalez, L.; Calvete, E. Design and
Measurement Properties of the Online Gambling Disorder Questionnaire (OGD-Q) in Spanish Adolescents. J. Clin. Med. 2020, 9,
15. [CrossRef]
37. Fung, S. Cross-cultural validation of the Social Media Disorder scale. Psychol. Res. Behav. Manag. 2019, 12, 683–690. [CrossRef]
38. Fargues, M.B.; Lusar, A.C.; Jordania, C.G.; Sanchez, X.C. Validation of two brief scales for Internet addiction and mobile phone
problem use. Psicothema 2009, 21, 480–485.
39. Wery, A.; Billieux, J. Problematic cybersex: Conceptualization, assessment, and treatment. Addict. Behav. 2017, 64, 238–246.
[CrossRef]
40. Wery, A.; Burnay, J.; Karila, L.; Billieux, J. The Short French Internet Addiction Test Adapted to Online Sexual Activities: Validation
and Links With Online Sexual Preferences and Addiction Symptoms. J. Sex Res. 2016, 53, 701–710. [CrossRef]
41. Davis, R.A. A cognitive-behavioral model of pathological Internet use. Comput. Hum. Behav. 2001, 17, 187–195. [CrossRef]
42. Caplan, S.E. Problematic Internet use and psychosocial well-being: Development of a theory-based cognitive-behavioral
measurement instrument. Comput. Hum. Behav. 2002, 18, 553–575. [CrossRef]
43. Douglas, A.C.; Mills, J.E.; Niang, M.; Stepchenkova, S.; Byun, S.; Ruffini, C.; Lee, S.K.; Loutfi, J.; Lee, J.K.; Atallah, M.; et al.
Internet addiction: Meta-synthesis of qualitative research for the decade 1996–2006. Comput. Hum. Behav. 2008, 24, 3027–3044.
[CrossRef]
44. Ozcan, N.K.; Buzlu, S. Internet use and its relation with the psychosocial situation for a sample of university students. CyberPsychol.
Behav. 2007, 10, 767–772. [CrossRef] [PubMed]
45. Wang, H.Y.; Zhao, Q.X.; Mu, W.T.; Rodriguez, M.; Qian, M.Y.; Berger, T. The Effect of Shame on Patients with Social Anxiety
Disorder in Internet-Based Cognitive Behavioral Therapy: Randomized Controlled Trial. JMIR Ment. Health 2020, 7, 12. [CrossRef]
46. Mahapatra, S. Smartphone addiction and associated consequences: Role of loneliness and self-regulation. Behav. Inf. Technol.
2019, 38, 833–844. [CrossRef]
47. Kannan, B.; Karthik, S.; Pal, G.K.; Menon, V. Gender Variation in the Prevalence of Internet Addiction and Impact of Internet
Addiction on Reaction Time and Heart Rate Variability in Medical College Students. J. Clin. Diagn. Res. 2019, 13, CC1–CC4.
[CrossRef]
48. Senol-Durak, E.; Durak, M. The Mediator Roles of Life Satisfaction and Self-Esteem between the Affective Components of
Psychological Well-Being and the Cognitive Symptoms of Problematic Internet Use. Soc. Indic. Res. 2011, 103, 23–32. [CrossRef]
49. Vally, Z. Generalized problematic Internet use, depression, and explicit self-esteem: Evidence from the United Arab Emirates.
Neurol. Psychiatry Brain Res. 2019, 33, 93–100. [CrossRef]
50. de la Villa Moral-Jiménez, M.; Fernández-Domínguez, S. Problematic internet use in spanish adolescents and their relationship
with self-esteem and impulsivity. Av. Psicol. Latinoam. 2019, 37, 103–119.
51. Kim, B.-N.; Park, S.; Park, M.-H. The relationship of sexual abuse with self-esteem, depression, and problematic internet use in
Korean adolescents. Psychiatry Investig. 2017, 14, 372. [CrossRef] [PubMed]
52. Mamun, M.A.; Hossain, M.S.; Moonajilin, M.S.; Masud, M.T.; Misti, J.M.; Griffiths, M.D. Does loneliness, self-esteem and
psychological distress correlate with problematic internet use? A Bangladeshi survey study. Asia Pac. Psychiatry 2020, 12, e12386.
[CrossRef]
53. Mathew, P. Impact of problematic internet use on the self-esteem of adolescents in the selected school, Kerala, India. Arch. Psychiatr.
Nurs. 2020, 34, 122–128. [CrossRef] [PubMed]
54. Li, C.; Liu, D.; Dong, Y. Self-esteem and problematic smartphone use among adolescents: A moderated mediation model of
depression and interpersonal trust. Front. Psychol. 2019, 10, 9. [CrossRef] [PubMed]
55. Elliott, G.C. Self-Esteem and Self-consistency–A theoretical and empirical link between 2 primary motivations. Soc. Psychol. Q.
1986, 49, 207–218. [CrossRef]
56. Grzegolowskaklarkowska, H.; Zolnierczyk, D. Defense of Self-esteem, defense of Self-consistency–A new voice in an old
controversy. J. Soc. Clin. Psychol. 1988, 6, 171–179. [CrossRef]
57. Gold, M. Scholastic experiences, Self-esteem and Delinquent-behavior—Theory for alternative schools. Crime Delinq. 1978, 24,
290–308. [CrossRef]
Int. J. Environ. Res. Public Health 2021, 18, 6949 12 of 14

58. Rosenberg, M.; Simmons, R.G. Black and White Self-Esteem: The Urban School Child; American Sociological Assocaition: Washington,
DC, USA, 1971.
59. Luthar, S.S.; Crossman, E.J.; Small, P.J. Resilience and adversity. In Handbook of Child Psychology and Developmental Science; Wiley:
New York, NY, USA, 2015; pp. 1–40.
60. Orth, U.; Robins, R.W.; Widaman, K.F. Life-span development of self-esteem and its effects on important life outcomes. J. Personal.
Soc. Psychol. 2012, 102, 1271. [CrossRef] [PubMed]
61. Rosenberg, M.; Schooler, C.; Schoenbach, C. Self-Esteem and Adolescent Problems: Modeling Reciprocal Effects. Am. Sociol. Rev.
1989, 54, 1004–1018. [CrossRef]
62. Pyszczynski, T.; Solomon, S.; Greenberg, J.; Arndt, J.; Schimel, J. Why do people need self-esteem? A theoretical and empirical
review. Psychol. Bull. 2004, 130, 435–468. [CrossRef] [PubMed]
63. Diener, E.; Diener, M. Cross-cultural correlates of life satisfaction and self-esteem. J. Personal. Soc. Psychol. 1995, 68, 653–663.
[CrossRef]
64. Rossi, A.; Panzeri, A.; Pietrabissa, G.; Manzoni, G.M.; Castelnuovo, G.; Mannarini, S. The Anxiety-Buffer Hypothesis in the Time
of COVID-19: When Self-Esteem Protects from the Impact of Loneliness and Fear on Anxiety and Depression. Front. Psychol.
2020, 11, 16. [CrossRef]
65. Fang, X.Y.; Zhang, J.Y.; Teng, C.; Zhao, K.; Su, K.P.; Wang, Z.; Tang, W.; Zhang, C. Depressive symptoms in the front-line
non-medical workers during the COVID-19 outbreak in Wuhan. J. Affect. Disord. 2020, 276, 441–445. [CrossRef] [PubMed]
66. Bozani, V.; Drydakis, N.; Sidiropoulou, K.; Harvey, B.; Paraskevopoulou, A. Workplace positive actions, trans people’s self-esteem
and human resources’ evaluations. Int. J. Manpow. 2020, 41, 809–831. [CrossRef]
67. Arsandaux, J.; Montagni, I.; Macalli, M.; Bouteloup, V.; Tzourio, C.; Galera, C. Health Risk Behaviors and Self-Esteem Among
College Students: Systematic Review of Quantitative Studies. Int. J. Behav. Med. 2020, 27, 142–159. [CrossRef] [PubMed]
68. Bozoglan, B.; Demirer, V.; Sahin, I. Loneliness, self-esteem, and life satisfaction as predictors of Internet addiction: A cross-sectional
study among Turkish university students. Scand. J. Psychol. 2013, 54, 313–319. [CrossRef]
69. Park, S.; Kang, M.; Kim, E. Social relationship on problematic Internet use (PIU) among adolescents in South Korea: A moderated
mediation model of self-esteem and self-control. Comput. Hum. Behav. 2014, 38, 349–357. [CrossRef]
70. Sechi, C.; Loi, G.; Cabras, C. Addictive internet behaviors: The role of trait emotional intelligence, self-esteem, age, and gender.
Scand. J. Psychol. 2020, 62, 9.
71. Chen, H.C.; Wang, J.Y.; Lin, Y.L.; Yang, S.Y. Association of Internet Addiction with Family Functionality, Depression, Self-Efficacy
and Self-Esteem among Early Adolescents. Int. J. Environ. Res. Public Health 2020, 17, 13. [CrossRef]
72. Dembinska, A.; Klosowska, J.; Ochnik, D. Ability to initiate relationships and sense of loneliness mediate the relationship between
low self-esteem and excessive internet use. Curr. Psychol. 2020, 1–7. [CrossRef]
73. Mo, P.K.H.; Chan, V.W.Y.; Wang, X.; Lau, J.T.F. Gender difference in the association between internet addiction, self-esteem and
academic aspirations among adolescents: A structural equation modelling. Comput. Educ. 2020, 155, 9. [CrossRef]
74. Geng, J.Y.; Han, L.; Gao, F.Q.; Jou, M.; Huang, C.C. Internet addiction and procrastination among Chinese young adults: A
moderated mediation model. Comput. Hum. Behav. 2018, 84, 320–333. [CrossRef]
75. Baturay, M.H.; Toker, S. Self-Esteem Shapes the Impact of GPA and General Health on Facebook Addiction: A Mediation Analysis.
Soc. Sci. Comput. Rev. 2017, 35, 555–575. [CrossRef]
76. Hou, Y.B.; Xiong, D.; Jiang, T.L.; Song, L.L.; Wang, Q. Social media addiction: Its impact, mediation, and intervention. Cyberpsy-
chology 2019, 13, 17. [CrossRef]
77. Zhang, Y.; Mei, S.L.; Li, L.; Chai, J.X.; Li, J.M.; Du, H.Y. The Relationship between Impulsivity and Internet Addiction in Chinese
College Students: A Moderated Mediation Analysis of Meaning in Life and Self-Esteem. PLoS ONE 2015, 10, e0131597. [CrossRef]
[PubMed]
78. Wang, P.C.; Wang, X.Y.; Wu, Y.Q.; Xie, X.C.; Wang, X.C.; Zhao, F.Q.; Ouyang, M.K.; Lei, L. Social networking sites addiction and
adolescent depression: A moderated mediation model of rumination and self-esteem. Pers. Individ. Differ. 2018, 127, 162–167.
[CrossRef]
79. Peng, W.Y.; Li, D.P.; Li, D.L.; Jia, J.C.; Wang, Y.H.; Sun, W.Q. School disconnectedness and Adolescent Internet Addiction:
Mediation by self-esteem and moderation by emotional intelligence. Comput. Hum. Behav. 2019, 98, 111–121. [CrossRef]
80. Celik, C.B.; Odaci, H. The relationship between problematic internet use and interpersonal cognitive distortions and life
satisfaction in university students. Child. Youth Serv. Rev. 2013, 35, 505–508. [CrossRef]
81. Kim, M.; Jun, H. The longitudinal effects of internet use on depression among korean older adults: Roles of employment and
social relationship satisfaction. Gerontologist 2015, 55, 476.
82. Okur, S.; Ozekes, M. Relationship between Basic Psychological Needs and Problematic Internet Use of Adolescents: The Mediating
Role of Life Satisfaction. Addicta 2020, 7, 214–222. [CrossRef]
83. Jeong, S.H.; Kim, H.; Yum, J.Y.; Hwang, Y. What type of content are smartphone users addicted to?: SNS vs. games. Comput. Hum.
Behav. 2016, 54, 10–17. [CrossRef]
84. Winthorst, W.H.; Bos, E.H.; Roest, A.M.; de Jonge, P. Seasonality of mood and affect in a large general population sample.
PLoS ONE 2020, 15, e0239033. [CrossRef]
85. Bailey, N.W.; Nguyen, J.; Bialylew, E.; Corin, S.E.; Gilbertson, T.; Chambers, R.; Fitzgerald, P.B. Effect on Well-Being from an
Online Mindfulness Intervention: Mindful in May. Mindfulness 2018, 9, 1637–1647. [CrossRef]
Int. J. Environ. Res. Public Health 2021, 18, 6949 13 of 14

86. Omar, S.Z.; Zaremohzzabieh, Z.; Abu Samah, A.; Bolong, J.; Shaffril, H.A.M. Influence of Different Facets of Internet Addiction on
Subjective Well-being in Malaysia: A Comparison Across Ethnic Groups. J. Komun. 2020, 36, 196–211. [CrossRef]
87. Schimmenti, A.; Starcevic, V.; Gervasi, A.M.; Deleuze, J.; Billieux, J. Interference with Processing Negative Stimuli in Problematic
Internet Users: Preliminary Evidence from an Emotional Stroop Task. J. Clin. Med. 2018, 7, 9. [CrossRef]
88. Matthews, N.; Farnsworth, B.; Griffiths, M.D. A Pilot Study of Problem Gambling among Student Online Gamblers: Mood States
as Predictors of Problematic Behavior. CyberPsychol. Behav. 2009, 12, 741–745. [CrossRef]
89. Guadagno, R.E.; Rempala, D.M.; Murphy, S.; Okdie, B.M. What makes a video go viral? An analysis of emotional contagion and
Internet memes. Comput. Hum. Behav. 2013, 29, 2312–2319. [CrossRef]
90. Wood, R.T.; Griffiths, M.D. A qualitative investigation of problem gambling as an escape-based coping strategy. Psychol. Psychother.
Theory Res. Pract. 2007, 80, 107–125.
91. Baloğlu, M.; Şahin, R.; Arpaci, I. A review of recent research in problematic internet use: Gender and cultural differences.
Curr. Opin. Psychol. 2020. [CrossRef] [PubMed]
92. Su, W.; Han, X.; Jin, C.; Yan, Y.; Potenza, M.N. Are males more likely to be addicted to the internet than females? A meta-analysis
involving 34 global jurisdictions. Comput. Hum. Behav. 2019, 99, 86–100. [CrossRef]
93. Mei, S.; Yau, Y.H.; Chai, J.; Guo, J.; Potenza, M.N. Problematic Internet use, well-being, self-esteem and self-control: Data from a
high-school survey in China. Addict. Behav. 2016, 61, 74–79. [CrossRef] [PubMed]
94. Yu, L.; Recker, M.; Chen, S.; Zhao, N.; Yang, Q. The moderating effect of geographic area on the relationship between age, gender,
and information and communication technology literacy and problematic internet use. Cyberpsychol. Behav. Soc. Netw. 2018, 21,
367–373. [CrossRef]
95. Chi, X.; Hong, X.; Chen, X. Profiles and sociodemographic correlates of Internet addiction in early adolescents in southern China.
Addict. Behav. 2020, 106, 106385. [CrossRef]
96. Vadher, S.B.; Panchal, B.N.; Vala, A.U.; Ratnani, I.J.; Vasava, K.J.; Desai, R.S.; Shah, A.H. Predictors of problematic Internet use in
school going adolescents of Bhavnagar, India. Int. J. Soc. Psychiatry 2019, 65, 151–157. [CrossRef] [PubMed]
97. Do, H.N.; Onyango, B.; Prakash, R.; Tran, B.X.; Nguyen, Q.N.; Nguyen, L.H.; Nguyen, H.Q.T.; Nguyen, A.T.; Nguyen, H.D.; Bui,
T.P. Susceptibility and perceptions of excessive internet use impact on health among Vietnamese youths. Addict. Behav. 2020, 101,
105898. [CrossRef]
98. Vickers, N.J. Animal communication: When I’m calling you, will you answer too? Curr. Biol. 2017, 27, R713–R715. [CrossRef]
[PubMed]
99. Tudorel, O.I.; Vintilă, M.; Vlaicu, L.; Bălăut, ă, D.; Goian, C.; Rusu, A. Romanian version of the Internet Addiction Test: Psychometric
properties and cross-gender invariance. Int. J. Ment. Health Addict. 2019, 17, 234–246. [CrossRef]
100. Hassan, T.; Alam, M.M.; Wahab, A.; Hawlader, M.D. Prevalence and associated factors of internet addiction among young adults
in Bangladesh. J. Egypt. Public Health Assoc. 2020, 95, 3. [CrossRef] [PubMed]
101. Fernandez-Villa, T.; Alguacil Ojeda, J.; Almaraz Gomez, A.; Cancela Carral, J.M.; Delgado-Rodriguez, M.; Garcia-Martin, M.;
Jimenez-Mejias, E.; Llorca, J.; Jose Molina, A.; Ortiz Moncada, R.; et al. Uso problematico de internet en estudiantes universitarios:
Factores asociados y diferencias de genero Problematic Internet Use in University Students: Associated factors and differences of
gender. Adicciones 2015, 27, 265–275. [CrossRef] [PubMed]
102. Tateno, M.; Teo, A.R.; Ukai, W.; Kanazawa, J.; Katsuki, R.; Kubo, H.; Kato, T.A. Internet addiction, smartphone addiction, and
Hikikomori trait in Japanese young adult: Social isolation and social network. Front. Psychiatry 2019, 10, 455. [CrossRef] [PubMed]
103. Vally, Z.; Laconi, S.; Kaliszewska-Czeremska, K. Problematic internet use, psychopathology, defense mechanisms, and coping
strategies: A cross-sectional study from the United Arab Emirates. Psychiatr. Q. 2020, 91, 587–602. [CrossRef]
104. Costa, R.M.; Patrão, I.; Machado, M. Problematic internet use and feelings of loneliness. Int. J. Psychiatry Clin. Pract. 2019, 23,
160–162. [CrossRef] [PubMed]
105. Cao, H.; Sun, Y.; Wan, Y.; Hao, J.; Tao, F. Problematic Internet use in Chinese adolescents and its relation to psychosomatic
symptoms and life satisfaction. BMC Public Health 2011, 11, 802. [CrossRef]
106. Watson, D.; Clark, L.A.; Tellegen, A. Development and validation of brief measures of positive and negative affect: The PANAS
scales. J. Personal. Soc. Psychol. 1988, 54, 1063–1070. [CrossRef]
107. Liu, J.D.; You, R.H.; Liu, H.; Chung, P.K. Chinese version of the international positive and negative affect schedule short form:
Factor structure and measurement invariance. Health Qual. Life Outcomes 2020, 18, 8. [CrossRef] [PubMed]
108. Liang, Y.; Zhu, D.M. Subjective Well-Being of Chinese Landless Peasants in Relatively Developed Regions: Measurement Using
PANAS and SWLS. Soc. Indic. Res. 2015, 123, 817–835. [CrossRef]
109. Mattson, R.E.; Rogge, R.D.; Johnson, M.D.; Davidson, E.K.; Fincham, F.D. The positive and negative semantic dimensions of
relationship satisfaction. Pers. Relatsh. 2013, 20, 328–355. [CrossRef]
110. Wong, D.S.W.; Fung, S.F. Development of the Cybercrime Rapid Identification Tool for Adolescents. Int. J. Environ. Res. Public
Health 2020, 17, 13. [CrossRef]
111. Mattson, R.E.; Lofgreen, A.M.; O’Farrell, T.J. Dyadic alcohol use, alcohol-specific conflict, and relationship dissatisfaction in
treatment-seeking men and their female partners. J. Soc. Pers. Relatsh. 2017, 34, 1206–1226. [CrossRef]
112. Wu, Y.; Zuo, B.; Wen, F.F.; Yan, L. Rosenberg Self-Esteem Scale: Method Effects, Factorial Structure and Scale Invariance Across
Migrant Child and Urban Child Populations in China. J. Personal. Assess. 2017, 99, 83–93. [CrossRef]
Int. J. Environ. Res. Public Health 2021, 18, 6949 14 of 14

113. Chen, W.; Luo, J.; Zhao, S.Y. Comparison of Two Chinese Revisions of the Rosenberg Self-Esteem Scale using Rasch Model. Int. J.
Psychol. 2016, 51, 943–944.
114. Xu, M.L.; Leung, S.O. Bifactor Structure for the Categorical Chinese Rosenberg Self-Esteem Scale. Span. J. Psychol. 2016, 19, 11.
[CrossRef]
115. West, S.G.; Finch, J.F.; Curran, P.J. Structural equation models with nonnormal variables: Problems and remedies. In Structural
Equation Modeling: Concepts, Issues, and Applications; Sage Publications, Inc.: Thousand Oaks, CA, USA, 1995; pp. 56–75.
116. Jöreskog, K.G. A general approach to confirmatory maximum likelihood factor analysis. Psychometrika 1969, 34, 183–202.
[CrossRef]
117. Bentler, P.M.; Bonett, D.G. Significance tests and goodness of fit in the analysis of covariance structures. Psychol. Bull. 1980, 88,
588. [CrossRef]
118. Kline, R.B. Principles and Practice of Structural Equation Modeling, 2nd ed.; Guilford Press: New York, NY, USA, 2005.
119. Schreiber, J.B.; Nora, A.; Stage, F.K.; Barlow, E.A.; King, J. Reporting Structural Equation Modeling and Confirmatory Factor
Analysis Results: A Review. J. Educ. Res. 2006, 99, 323–338. [CrossRef]
120. Hu, L.T.; Bentler, P.M. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives.
Struct. Equ. Model. A Multidiscip. J. 1999, 6, 1–55. [CrossRef]
121. Preacher, K.J.; Rucker, D.D.; Hayes, A.F. Addressing Moderated Mediation Hypotheses: Theory, Methods, and Prescriptions.
Multivar. Behav. Res. 2007, 42, 185–227. [CrossRef] [PubMed]
122. Muthén, L.; Muthén, B. Mplus, User’s Guide, 8th ed.; Muthén & Muthén: Los Angeles, CA, USA, 2017.
123. Wald, A. Tests of statistical hypotheses concerning several parameters when the number of observations is large. Trans. Am. Math.
Soc. 1943, 54, 426–482. [CrossRef]
124. Kim, D.; Lee, Y.; Lee, J.; Nam, J.K.; Chung, Y. Development of Korean Smartphone Addiction Proneness Scale for Youth. PLoS
ONE 2014, 9, e97920. [CrossRef] [PubMed]
125. Caplan, S.E. Theory and measurement of generalized problematic Internet use: A two-step approach. Comput. Hum. Behav. 2010,
26, 1089–1097. [CrossRef]