Problematic internet use profiles and psychosocial risk among adolescents

Halley M. Pontes; Mirna Macur

doi:10.1371/journal.pone.0257329

Abstract

Objective

Although Problematic Internet Use (PIU) is an emerging area of study in psychology, little is known about the unique features of specific subgroups of internet users and their psychosocial vulnerabilities within robust and nationwide populations.

Methods

The aim of this study was to identify distinct latent groups of internet users based on their PIU risk and to compare their psychosocial outcomes. To achieve this, a nationally representative sample of adolescents of the same grade (N = 1,066, Mean_age = 13.46 years, range = 12–16) was recruited from several schools in Slovenia through stratified random sampling.

Results

A Latent Profile Analysis (LPA) revealed a two-class solution, with Class 1 (n = 853, 80%) featuring ‘low PIU risk’ participants and Class 2 (n = 213, 20%) including ‘high PIU risk’ participants. Behaviorally, the main feature of Class 1 denoted ‘time management difficulties’ while Class 2 was best characterized by ‘mood and time management issues’. Further frequentist and Bayesian analyses indicated that Class 2 presented greater psychosocial risk compared to Class 1 due to significantly higher levels of PIU (generalized and across specific PIU subfactors) coupled with lower levels of subjective well-being and self-control.

Conclusions

Contrary to what was initially envisaged, the two classes did not differ in terms of perceived quality in parent-child relationship. This study shows that PIU patterns and symptom-severity may be developmentally specific, further highlighting the need for clinically age-adjusted PIU screening practices within epidemiological and healthcare settings.

Citation: Pontes HM, Macur M (2021) Problematic internet use profiles and psychosocial risk among adolescents. PLoS ONE 16(9): e0257329. https://doi.org/10.1371/journal.pone.0257329

Editor: Chung-Ying Lin, National Cheng Kung University College of Medicine, TAIWAN

Received: February 16, 2021; Accepted: August 27, 2021; Published: September 14, 2021

Copyright: © 2021 Pontes, Macur. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: Interested readers are encouraged to forward enquires about the data set directly to the National Institute of Public Health of Slovenia by postal correspondence (Trubarjeva 2, 1000 Ljubljana, Slovenia), telephone (+386 1 2441 400), or by email (info@nijz.si). We confirm that the authors received no special privileges in accessing the data and that the National Institute of Public Health of Slovenia did not influence any aspect of the present research as the study was conducted independently by the authors.

Funding: We acknowledge that this research has received financial support from Norway Grants SI05-0007 and was conducted in partnership with the National Institute of Public Health of Slovenia and the Slovenian youth association No Excuse, managed by Dr. Mirna Macur.

Competing interests: The authors have declared that no competing interests exist.

Introduction

The latest statistics suggest that approximately 4.66 billion individuals worldwide actively use the internet, representing a total global penetration rate of 59% [1]. Given the multiple functionalities and services offered by the internet as a contemporary leisure and professional tool, its use has been found to lead to several psychosocial benefits [2], with social media use constituting a constructive coping strategy capable of helping internet users manage their loneliness and anxiety levels during unprecedented pandemic times [3], and healthy forms (e.g., harmonious as opposed to obsessive passion) of online gaming engagement being associated with increased bonding capital, decreased loneliness, and enhanced well-being [4].

Theoretically, these benefits can be understood within the Needs-Affordances-Features perspective [5] on technology (e.g., internet) use, which suggests that users’ psychological needs drive their usage levels of specific digital technologies and features (e.g., social media, gaming, gambling, emailing, pornography, shopping, etc.), to the extent that such usage provides salient affordances that ultimately contribute to satisfying their basic psychological needs.

Notwithstanding this, the perils associated to excessive and dysregulated use of the internet have been widely documented within the psychological literature [6], with the first studies on Problematic Internet Use (PIU), also controversially termed as ‘Internet addiction’ [7,8], being published nearly 25 years ago [see 9,10]. Although, generalized PIU is not currently recognized as an official mental health disorder, gaming disorder [seen as a specific form of PIU, see 11], has been recently recognized as a bona fide addictive disorder [12], and previously proposed as a tentative disorder by the American Psychiatric Association (APA) under the term of ‘Internet Gaming Disorder’ [13].

In relation to the epidemiological status of PIU, a number of key review studies reported important findings. A recent meta-analytic study reviewing 113 studies from 31 countries suggested a prevalence rate of PIU of 7%, with prevalence increasing over time and differing according to the assessment tool utilized [14]. In terms of gender-based prevalence, a recent meta-analysis of 101 studies from 34 nations [15] reported that male internet users presented with higher PIU tendency compared to females, with the largest effect sizes in gender-based prevalence estimates being observed in Asia and the lowest in North America. As for geographical prevalence estimates, a meta-analysis of 80 studies from 31 countries covering seven world regions found a global prevalence of PIU of 6%, with the highest prevalence being observed in the Middle East region (10.9%), followed by North America (8%), and Southern and Eastern Europe (6.1%). Although epidemiology appears to vary across different regions, PIU vulnerability is relatively equivalent across Western and non-Western nations [14].

Several studies have also widely documented the detrimental psychosocial impacts of PIU. A recent study investigating cognitive performance in PIU concluded that the existing evidence supports a significant association between PIU and decrements across a range of neuropsychological domains (e.g., inhibitory control, decision-making, and working memory) [16]. Further investigation indicated that PIU may result in sleep problems (i.e., decreased sleep quality and duration) [17] and decreased perceived social support [18].

Recently, Fumero, Marrero [19] have investigated the role specific PIU risk factors linked to psychopathology (e.g., depression and anxiety), personality (e.g., aggressiveness, hostility, and sensation seeking self-esteem), and social factors (e.g., peer problems and family difficulties), and found that emotional personal factors (e.g., hostility, depression, and anxiety) had greater associations with PIU than social factors, with PIU being mainly associated with mood disturbances. In terms of comorbidities, PIU has been linked to greater levels of somatic symptoms, clinical depression, psychoticism, paranoid ideation, serious mental illness, and suicidality (i.e., ideations, plans, and attempts) [20], with additional research supporting a consistent association between PIU and a wide range of adverse mental health outcomes such as attention deficit/hyperactivity disorder [21], social anxiety [22], and alcohol abuse [23].

Despite its current unofficial clinical status, PIU affects a significant, albeit relatively small, portion of the population worldwide, further warranting additional scientific scrutiny on its potential impacts, especially within large-scale and nationwide studies that are able to provide more generalizable estimates regarding the pitfalls of PIU.

The present study

The main aim of the present study was twofold: (i) to identify and characterize unique subgroups of internet users based on their PIU risk in order to further (ii) compare their distinct psychosocial differences. To achieve this, we have opted to conduct a Latent Profile Analysis (LPA), because this analytical approach is usually employed to understand variables (or indicators) that can predict high-risk behaviors through identification of latent subgroups presenting varying risk levels [24,25].

This statistical approach offers important advantages over more traditional dimensional approaches (e.g., factor analytical) because it does not assume that the probability of a response on a putative variable is a linear function of a latent factor, instead, the variables used in the LPA predict class membership based on the intersection of variables used in the model defining subgroups that differ in qualitative terms [26]. Ultimately, this investigatory approach aligns with the latest recommendations to investigate problematic technology use behaviors employing non-confirmatory approaches [27]. Thus, in light of the existing literature and the aim established for this study, the two following exploratory working hypotheses were devised:

Hypothesis 1 (H1): specific latent subgroups among internet users will be identified based on their unique PIU patterns and risk. We expect that a unique typology of participants will emerge based on their intricate shared PIU risks.

Hypothesis 2 (H2): the latent subgroups of internet users identified in the LPA will differ in terms of their psychosocial risks whereby, when compared to participants showing lower risk, those experiencing higher risk will be expected to exhibit greater levels of PIU symptoms coupled with reduced levels of subjective well-being, quality in perceived parent-child relationship, and self-control levels.

Given the large amount of PIU studies using relatively low sample sizes recruited through convenience sampling, this study will contribute to the field by further helping build the knowledge base regarding the psychosocial risks associated with PIU within a large-scale and nationwide framework, generating findings that better reflect users’ experiences and psychosocial vulnerabilities.

Material and methods

Participants

Data for this study was collected nationwide in the context of a large epidemiological study in Slovenia. Participants were adolescents recruited via stratified random sampling using all primary school classes (i.e., eighth grade) as the primary sampling units. The stratification was based on the twelve statistical regions of Slovenia according to the Nomenclature of Territorial Units for Statistics (i.e., NUTS 3) and population density.

All school principals of the randomly sampled schools were contacted and invited to participate in the study. The study procedures were carried out in accordance with the Declaration of Helsinki. The Institutional Review Board of Nottingham Trent University (United Kingdom) and the National Institute of Public Health (Slovenia) approved the study. All subjects were informed about the study and they all provided written informed consent. Parental written informed consent was provided by parents and/or legal guardians of participants aged below 15 years.

A computer-assisted survey containing all main measures of the study was administered by trained staff and electronically completed by all participants in the school setting (April to May 2015). Prior to the data collection, informed consent was obtained from all parents and/or legal guardians, and a sample of 1,095 participants was initially recruited to the present study. All participants presenting with severe missing values (i.e., ≥ 10%, n = 24, 2.2%) and carelessness or inattentiveness when filling out the survey (i.e., ≥ 50% response invariability, n = 5, 0.5%) were excluded. Thus, the effective sample comprised 1,066 adolescents (Mean_age = 13.46 years, SD_age = 0.58) (see Table 1).

Download:

Table 1. Main sociodemographic characteristics of the sample (N = 1,066).

https://doi.org/10.1371/journal.pone.0257329.t001

Measures

Sociodemographic and internet use data.

Sociodemographic data were collected on participants’ age, gender, and region of residence. Moreover, internet use behaviors included assessing participants daily time spent on the internet for leisure purposes (i.e., number of hours) on weekdays and weekends in general and through their mobile phones more specifically.

Problematic internet use.

The Problematic Internet Use Questionnaire Short-Form (PIUQ-SF-6) [28] was used to assess PIU. This brief tool conceptualizes PIU according to three subfactors: Obsession (items 2: “How often do you feel tense, irritated, or stressed if you cannot use the Internet for as long as you want to?” and 6: “How often does it happen to you that you feel depressed, moody, or nervous when you are not on the Internet and these feelings stop once you are back online?”), Neglect (items 1: “How often do you spend time online when you’d rather sleep?” and 5: “How often do people in your life complain about you spending too much time online?”), and Control Disorder (items 3: “How often does it happen that you wish to decrease the amount of time spend online but you do not succeed?” and 4: “How often do you try to conceal the amount of time spent online?”). Each factor contains two items rated on a five-point Likert scale ranging from ‘never’ to ‘always/almost always’ and total obtainable scores can range from 6 to 30, with higher scores indicating greater levels of PIU. The PIUQ-SF-6 has been previously shown to be psychometrically sound and has been utilized across several studies [e.g., 29–35]. In the present study, the PIUQ-SF-6 exhibited excellent internal consistency levels (Cronbach’s α = .84; McDonald’s ω = .90).

Subjective well-being.

Participants’ subjective well-being was evaluated using a composite measure comprising the following two Likert-type self-report items: “Are you satisfied with your life?” (rated from 1 = ‘extremely dissatisfied’ to 10 = ‘extremely satisfied’) and “In general, how would you consider your mental health?” (rated from 1 = ‘bad’ to 5 = ‘excellent’). The two items were summed to represent an overall measure of participants’ subjective well-being, with higher scores indicating increased levels of perceived subjective well-being. In the present study, this composite measure exhibited excellent internal consistency levels (Cronbach’s α = .67; McDonald’s ω = .75).

Parent-child relationship.

The quality of parental relationship was assessed using four items measuring perceived levels of relational affection between the participants and their parents (e.g., “My parents and I try to spend much time together”). These four items have been previously used in similar research that supported its psychometric properties [36]. Moreover, each item represents different shared activities and levels of emotional closeness that are rated on a five-point Likert scale ranging from 1 = ‘very untrue’ to 5 = ‘very true’, with higher scores reflecting increased perceived quality in parental relationship. In the present study, this measure has exhibited excellent levels of internal consistency (Cronbach’s α = .85; McDonald’s ω = .87).

Self-control.

Participants’ level of self-control was evaluated using a brief seven-item measure that has been employed in previous PIU research supporting its adequate psychometric properties [36]. All items of this psychometric test (e.g., “I feel like I am a ticking time bomb”) are rated on a five-point Likert scale ranging from 1 = ‘strongly agree’ to 5 = ‘strongly disagree’, with higher scores indicating greater levels of self-control. In the present study, this measure has exhibited excellent internal consistency levels (Cronbach’s α = .82; McDonald’s ω = .86).

Statistical analysis

All statistical analyses were performed using the statistical packages R version 4.0.3 “Bunny-Wunnies Freak Out” [37] and Mplus version 8.2 [38]. The analytical approach adopted in this study included the three main steps: (i) basic descriptive statistical analysis (e.g., means, standard deviations) of the samples’ main features, (ii) LPA to identify specific typologies and latent subgroups (i.e., classes) of internet users based on their PIU risk, and (iii) inferential statistics (e.g., frequentist and Bayesian independent samples Welch’s t-tests) accompanied by effect size estimations (e.g., Cohen’s d) to better characterize the latent classes identified in the LPA.

While LPA was conducted with Mplus due to its ability to incorporate sampling weights in the mixture modeling procedure, all remaining analyses were performed on R using the packages careless version 1.1.3 [39], BayesFactor version 0.9.12–4.2 [40], psych version 2.0.9 [41], and weights version 1.0.1 [42] embedding sampling weights in the estimations. The estimated Bayes factors (BF₁₀) were interpreted according to the recommendations provided by Lee and Wagenmakers [43] and Cohen’s d effect sizes were interpreted as small (d = .2), medium (d = .5), and large (d = .8) [44].

The LPA is a mixture modeling statistical analysis that can be used to identify groups of individuals that may be similar in their response patterns to specific variables. In the presented study, LPA was conducted using all six PIUQ-SF-6 items (continuous variables) as opposed to using the three subfactors of the PIUQ-SF-6 since this approach offers enhanced analytical granularity in the description and interpretation of the latent classes due to the greater amount of variables informing the estimation procedure.

To determine the optimal number of latent classes to be extracted in the LPA, the following three Information Theory Fit Indices were inspected: Akaike’s Information Criterion (AIC) [45], Bayesian Information Criterion (BIC) [46], and the Sample-Size Adjusted Bayesian Information Criterion (SSABIC) [47], with lower values indicating better model fit [48]. Additionally, Entropy [49] was computed to ascertain whether the estimated solution was optimal. Entropy ranges from 0 to 1, and larger values suggest greater clarity in the identification of the latent classes while values approaching 1 indicate clear separation of the classes [50]. Furthermore, the adjusted Lo-Mendell-Rubin Likelihood Ratio Test (L-M-R Test) [51] was also computed to determine the optimal number of classes by using its p-value, with low and significant p-values (i.e., p < .05) suggesting that the K₀-class model has significantly better fit to the data than the K-₁-class model [48]. This likelihood-based test performs less well compared to the Bootstrap Likelihood Ratio Test (BLRT) [52], however, the BLRT cannot be computed when using sampling weights.

Finally, after the optimal number of latent classes underlying the data was identified, further comparative analyses were conducted to characterize the unique features of each class in relation to overall and specific PIU, subjective well-being, parent-child relationship, and self-control levels.

Results

Descriptive statistics

In terms of basic weighted sociodemographic features, the mean age of the sample was 13.46 years (SD_age = 0.58, range = 12–16 years) and the gender distribution was even (n = 533, 50% female). Table 1 provides a summary of all key sociodemographic features, including the sample’s residency statistics. In relation to internet use behaviors, participants reported spending greater amounts of time on the internet for leisure purposes on the weekends (Mean = 3.34 hours, SD = 2.87), with mobile internet use being mostly prominent on the weekends (Mean = 1.99 hours, SD = 2.44).

LPA

Determining the number of latent classes.

The LPA was performed using specified starting values with random starts (2,000 random sets of starting values for the initial stage and 20 final stage optimizations) to avoid solutions at local maxima, using maximum likelihood estimation with robust standard errors (MLR) as the estimator. The full results of the LPA for models with 1 through 4 classes are presented in Table 2. After evaluating all four models, the two-class model was chosen since it provided the best fit to the data.

Download:

Table 2. Model fit of the latent profile analysis models from 1 through 4 classes (N = 1,066).

https://doi.org/10.1371/journal.pone.0257329.t002

The decision to favor the two-class model was based on the fact that the L-M-R Test was not statistically significant for the three- (p = .45) or four-class (p = .13) models, with entropy being optimal for the two-class model (i.e., 0.935). Furthermore, the two-class model was able to correctly classify 98.8% of Class 1 and 95.4% of Class 2 participants, which is well-above the recommended threshold of .70 [53]. Although the three- and four-class models had lower BIC, AIC, and SSABIC values, the three-class model was disregarded as it presented with (i) less optimal L-M-R Test p-values (i.e., > .05), (ii) worse accuracy in class membership assignment due to lower entropy levels, and (iii) less theoretical clarity in interpreting the solution derived. Based on current LPA conventions [54], the alternative four-class model was also disregarded as it did not converge.

Latent class characterization and interpretation.

The two final classes that emerged from the LPA alongside their main characteristics in terms of response patterns to the PIUQ-SF-6 items (i.e., mean PIU severity on each item) is illustrated in Fig 1. The two classes were labelled relative to their problematic usage risk in order to add qualitative value to the interpretability of the LPA results as commonly done in similar studies [e.g., 55–59]. Overall, Class 1 (n = 853, 80%) presented lower PIU symptom-severity in comparison to Class 2 (n = 213, 20%). Thus, Class 1 was labelled as ‘low PIU risk’ while Class 2 was labelled as ‘high PIU risk’.

Download:

Fig 1. Characterization of the two latent classes of problematic internet users obtained from the Latent Profile Analysis (N = 1,066).

Note. Problematic internet use symptom means for each item of the Problematic Internet Use Questionnaire Short-Form (PIUQ-SF-6). PIUQ1 = How often do you spend time online when you’d rather sleep?, PIUQ2 = How often do you feel tense, irritated, or stressed if you cannot use the Internet for as long as you want to?; PIUQ3 = How often does it happen that you wish to decrease the amount of time spent online but you do not succeed?; PIUQ4 = How often do you try to conceal the amount of time spent online?; PIUQ5 = How often do people in your life complain about you spending too much time online?; PIUQ6 = How often does it happen to you that you feel depressed, moody, or nervous when you are not on the Internet and these feelings stop once you are back online?.

https://doi.org/10.1371/journal.pone.0257329.g001

In terms of class characterization, ‘low PIU risk’ participants presented particularly high symptoms related to the PIU subfactors associated with Control Disorder (item 3) and Neglect (items 5 and 1), clearly indicating that the main behavioral feature of this class is marked by time management difficulties. In contrast, ‘high PIU risk’ participants exhibited elevated symptoms related to the Obsession (item 2), Control Disorder (item 3), and Neglect (item 1) subfactors, suggesting that the main behavioral aspect of this class is marked by mood and time management issues.

Cross-class comparisons.

In addition to the aforementioned procedures, the two classes were further compared against each other with respect to participants’ overall levels of PIU and its three specific subfactors, alongside subjective well-being, parent-child relationship, and self-control levels (see Table 3).

Download:

Table 3. Cross-class comparisons of high and low risk of problematic internet users for the two latent classes (N = 1,066).

https://doi.org/10.1371/journal.pone.0257329.t003

Accordingly, in comparison to the ‘low PIU risk’ participants (i.e., Class 1), participants classed as having ‘high PIU risk’ (i.e., Class 2) presented with significantly higher overall PIU symptoms (t_[275.48] = -12.24, p < .001; BF₁₀ = 67, d = 1.05) and across the three PIU subfactors: Obsession (t_[255.38] = -11.07, p < .001; BF₁₀ = 55, d = 1.06), Neglect (t_[278.95] = -10.25, p < .001; BF₁₀ = 47, d = 0.84), and Control Disorder (t_[287.60] = -7.64, p < .001; BF₁₀ = 25, d = 0.66).

Additionally, as expected, in comparison to ‘low PIU risk’ participants, those with membership to the ‘high PIU risk’ class exhibited significantly lower levels of subjective well-being (t_[309.36] = 2.91, p = .004; BF₁₀ = 1.70, d = 0.28) and self-control (t_[596.09] = 25.51, p < .001; BF₁₀ > 100, d = 0.24). Contrary to what has been hypothesized, parent-child relationship did not differ significantly between the two classes (t_[330.10] = 1.79, p = .074; BF₁₀ = 0.89, d = 0.14). The magnitude of the observed effect sizes across the means compared ranged from small to large. Taken together, these results lend support to H1, and partially support H2.

Discussion

The main aim of this study was to help advance the literature on PIU by exploring its psychosocial risks in a nationally representative sample of adolescents. This constitutes an important goal given that most previous research on PIU focused on investigating older populations (e.g., emerging adults and adults). This study allowed us to empirically identify and characterize unique subgroups of internet users based on their unique PIU risk so that we could estimate their potential psychosocial risks based on group membership. Below, we discuss the implications of the findings obtained according to the existing literature and the goals of the present study.

In the context of H1, we expected that specific subgroups of adolescent internet users would emerge based on their PIU risk. Although previous research has performed Latent Class Analysis (LCA) to investigate this issue, these studies have employed different assessment methodologies using the existing alternative forms of the PIUQ assessment tool [60]. To the best of our knowledge, only one study has utilized the PIUQ-SF-6 to identify latent classes [28].

Using a nationally representative sample of Hungarian adolescents (mean age of 16.4 years), Demetrovics, Király (28) also found two distinct latent classes based on the three subfactors of the PIUQ-SF-6. More specifically, their ‘at risk of PIU’ class (comprising 14.4% of the sample) was mainly characterized by elevations on Neglect (items 1 and 5), Obsession (items 2 and 6), and Control Disorder (items 3 and 4). Based on the LPA findings of our study, ‘high PIU risk’ participants exhibited a slightly different (but still comparable) pattern of PIU risk as this class was mainly characterized by the one item of the Obsession (i.e., item 2), Control Disorder (i.e., item 3), and the Neglect (i.e., item 1) subscales.

While ‘at risk of PIU’ participants from the study conducted by Demetrovics, Király [28] were mainly characterized in terms of the symptoms measured by items 1 and 5, in our study the most pronounced features of the ‘high PIU risk’ class related to symptoms measured by items 1, 2, and 3. Although there are some overlapping features between these two groups (i.e., item 1), our ‘high PIU risk’ class was proportionally larger (i.e., 20%) and experienced more diffuse PIU symptoms beyond the Neglect subfactor.

This nuanced and seemingly discrepant finding might underlie unique developmental features and parental mediation approaches adopted by parents to regulate their child’s internet use behaviors. It is known that parental mediation is a dynamic process whereby parents adapt their mediation styles to their developing child while also adjusting their internet mediation styles to their child’s stage of development [61,62]. Thus, it is plausible that internet-related parental mediation strategies may have differed between children and adolescent internet users from our study and the study conducted by Demetrovics, Király [28].

We note that although parents may employ developmentally-adjusted mediation strategies when regulating their child’s internet use, there is currently no robust evidence supporting a strong link between specific parental mediation strategies and decreased PIU levels [63]. However, it is likely that different mediation strategies may lead to distinct outcomes in relation to internet use behavior and its many resulting effects. Relatedly, psychoeducational interventions may also be fruitful in raising awareness about the detrimental effects of PIU by fostering healthy and judicious usage of the internet, preventing problematic usage.

In relation to the latent subgroups of internet users presenting diminished PIU risk, Demetrovics, Király [28] found a second class that was labelled as ‘not at risk of PIU’ (comprising 85.6% of the sample), which was mostly characterized by PIUQ-6-SF items measuring Neglect. This finding is highly consistent with the present findings as our ‘low PIU risk’ class was proportionally equivalent in size (i.e., 80%) and similar in terms of the most pronounced PIU symptoms experienced.

In addition to determining and characterizing the latent PIU classes reported in this study, we also sought to compare these classes in order to establish participants’ unique psychosocial risk profiles. Therefore, based on the results obtained, we found partial support to H2. Although participants with membership to the ‘high PIU risk’ class had greater levels of PIU (as an overall measure and across its three subfactors), lower subjective well-being, and self-control, our results did not support the expected assumption that ‘high PIU risk’ participants would experience lower levels of quality in parent-child relationship. There may be several potential reasons underpinning the lack of a significant relationship between greater PIU risk and lower quality in parent-child relationship.

Firstly, PIU is not yet an officially recognized and psychiatrically accepted addictive disorder. In fact, the notion of ‘Internet addiction’ remains highly controversial and its clinical validity and usefulness has been debated for many years [7,8,64]. While the clinical status of PIU is currently being scientifically challenged, it is clear that if PIU (in its generalized and non-specific form) was to be a bona fide addictive disorder, the expected relationships in our cross-class comparisons would likely have been corroborated. Secondly, the psychometric tool used in this study to assess PIU (i.e., PIUQ-6-SF) is an extremely brief tool, and its relatively low number of items may not necessarily provide optimal coverage of the six core components of addictive disorder underpinning higher PIU severity [65].

The relationship between parental relationship and PIU is complex and multifaceted. For example, in a slightly older sample of adolescents, Wang, Li [66] reported that greater quality in parent-adolescent relationship was negatively associated with PIU, supporting the view that a favorable parental relationship may constitute a protective developmental resource capable of buffering problematic behaviors in adolescence. This is because a functional parent-adolescent relationship contributes to the development of emotion regulation skills that can decrease PIU levels [66]. In a recent similar study involving older adolescents, Huang, Hu [67] reported that the quality in the relationship between parent-child was negatively associated to PIU, with this association being mediated by self-concept. Furthermore, in the context of specific PIU problems related to online gaming, among children aged between 12 to 17 years, the quality in parent-child relationship has been reported as a weak correlate of internet gaming problems [68].

The findings of our study present with several important implications to different stakeholders (e.g., caretakers, practitioners, etc.). At the parental level, because high-risk PIU individuals are usually young [69–71], it is important that, beyond employing specific mediation styles, parents work with their adolescent to bolster their levels of emotion regulation as it constitutes a risk factor for PIU [70] exerting indirect effects on problematic usage as evidenced by several studies [66,72]. At the clinical level, if PIU is capable eliciting functional impairments, mental health practitioners should employ clinical interview methodologies and assessment approaches that evaluate adolescents’ mood and time management issues due to their dysregulated internet use as this may provide valuable information about the potential PIU risks and severity of internet-related functional impairments that may be experienced by younger clients.

Moreover, based on the findings obtained in our study and the items mostly predictive of high PIU risk, when it comes to PIU assessment in adolescents, mental health practitioners should pay particular attention to elevated mood-related issues (e.g., withdrawal symptoms, item 2), failure to control the behavior (e.g., internet use dysregulation, item 3), and internet overuse that negatively impacts on quality of life (e.g., sleep problems, item 1) as these were the main features presented by adolescents exhibiting greater risk of PIU in the present study.

The present study also paves the way to future research aiming at expanding the extant knowledge on PIU among youth. Future research may provide valuable contributions to the field by examining how clinically diagnosed PIU individuals compare non-PIU individuals across different psychological and social domains using different analytical approaches such as logistic regression modeling in order to ascertain the probability specific functional impairments may occur across different risk groups.

Despite these promising findings, our results are not free from potential limitations. Firstly, given the large-scale nature of this study and the financial costs involved, we were not able to measure additional factors that may help explain the intricacies between psychosocial factors and PIU risk. Therefore, our findings do not signify the existence of causal relationships based on the results reported as there may be additional underlying mechanisms explaining the relationship between elevated PIU symptoms and greater psychosocial risk. Secondly, the cross-sectional nature of our study hinders the ability to establish directionality for the effects observed even though the relationships reported are likely to take place in a sequenced way. Thirdly, the study would have been able to provide greater insights onto PIU and its associated risks by sampling young internet users from different age groups in order to provide a broader picture of how PIU may unfold across different developmental stages.

Notwithstanding these potential limitations, our findings provide a valuable contribution to the field by helping consolidate the current understanding about the potential impacts of PIU in childhood. The many strengths of this study include the use of a robust analytical and modeling framework to investigate the hypotheses developed, and the use of a large and nationally representative sample of a specific population that remains overlooked in PIU research (i.e., adolescents).

Conclusion

Based on the findings obtained, it can be concluded that about 20% of Slovenian youth (i.e., eighth graders) are at a greater risk for developing PIU symptoms and related problems. Moreover, greater PIU risk may translate to time management difficulties coupled with mood management issues. The findings obtained further indicated that greater PIU risk was associated with increase PIU symptomatology and reduced levels of subjective well-being and self-control.

References

1. Statista. Worldwide digital population as of October 2020 2020 [updated November 23November 27, 2020]. Available from: https://www.statista.com/statistics/617136/digital-population-worldwide.
2. Tifferet S. Gender Differences in Social Support on Social Network Sites: A Meta-Analysis. Cyberpsychol Behav Soc Netw. 2020;23(4):199–209. pmid:32096662
- View Article
- PubMed/NCBI
- Google Scholar
3. Cauberghe V, Van Wesenbeeck I, De Jans S, Hudders L, Ponnet K. How Adolescents Use Social Media to Cope with Feelings of Loneliness and Anxiety During COVID-19 Lockdown. Cyberpsychol Behav Soc Netw. 2020. pmid:33185488
- View Article
- PubMed/NCBI
- Google Scholar
4. Mandryk RL, Frommel J, Armstrong A, Johnson D. How Passion for Playing World of Warcraft Predicts In-Game Social Capital, Loneliness, and Wellbeing. Front Psychol. 2020;11(2165). pmid:33071843
- View Article
- PubMed/NCBI
- Google Scholar
5. Karahanna E, Xu SX, Xu Y, Zhang NA. The needs–affordances–features perspective for the use of social media. MIS Quarterly: Management Information Systems. 2018;42(3):737–56.
- View Article
- Google Scholar
6. Pontes HM, Kuss DJ, Griffiths MD. Clinical psychology of Internet addiction: A review of its conceptualization, prevalence, neuronal processes, and implications for treatment. Neuroscience and Neuroeconomics. 2015;4:11–23.
- View Article
- Google Scholar
7. Starcevic V, Aboujaoude E. Internet addiction: Reappraisal of an increasingly inadequate concept. CNS Spectrums. 2016;22(1):7–13. pmid:26831456
- View Article
- PubMed/NCBI
- Google Scholar
8. Starcevic V. Is Internet addiction a useful concept? Australian and New Zealand Journal of Psychiatry. 2013;47(1):16–9. pmid:23293309
- View Article
- PubMed/NCBI
- Google Scholar
9. Griffiths MD. Internet "addiction": an issue for clinical psychology? Clinical Psychology Forum. 1996;97:32–6.
- View Article
- Google Scholar
10. Young KS. Psychology of computer use: XL. Addictive use of the Internet: A case that breaks the stereotype. Psychological Reports. 1996;79(3):899–902. pmid:8969098
- View Article
- PubMed/NCBI
- Google Scholar
11. Young KS, Pistner M, O’Mara J, Buchanan J. Cyber disorders: The mental health concern for the new millennium. Cyberpsychol Behav. 1999;2(5):475–9. pmid:19178220
- View Article
- PubMed/NCBI
- Google Scholar
12. World Health Organization. International statistical classification of diseases and related health problems (11th ed.).2019.
13. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (5th ed.)2013.
14. Pan YC, Chiu YC, Lin YH. Systematic Review and Meta-Analysis of Epidemiology of Internet Addiction. Neuroscience & Biobehavioral Reviews. 2020;118:612–22. pmid:32853626
- View Article
- PubMed/NCBI
- Google Scholar
15. Su W, Han X, Jin C, Yan Y, Potenza MN. Are males more likely to be addicted to the internet than females? A meta-analysis involving 34 global jurisdictions. Comput Human Behav. 2019;99:86–100.
- View Article
- Google Scholar
16. Ioannidis K, Hook R, Goudriaan AE, Vlies S, Fineberg NA, Grant JE, et al. Cognitive deficits in problematic internet use: meta-analysis of 40 studies. British Journal of Psychiatry. 2019;215(5):639–46. Epub 2019/02/20. pmid:30784392
- View Article
- PubMed/NCBI
- Google Scholar
17. Alimoradi Z, Lin CY, Broström A, Bülow PH, Bajalan Z, Griffiths MD, et al. Internet addiction and sleep problems: A systematic review and meta-analysis. Sleep Medicine Reviews. 2019;47:51–61. pmid:31336284
- View Article
- PubMed/NCBI
- Google Scholar
18. Lei H, Li S, Chiu MM, Lu M. Social support and Internet addiction among mainland Chinese teenagers and young adults: A meta-analysis. Comput Human Behav. 2018;85:200–9.
- View Article
- Google Scholar
19. Fumero A, Marrero RJ, Voltes D, Peñate W. Personal and social factors involved in Internet addiction among adolescents: A meta-analysis. Comput Human Behav. 2018;86:387–400.
- View Article
- Google Scholar
20. Guo W, Tao Y, Li X, Lin X, Meng Y, Yang X, et al. Associations of Internet Addiction Severity With Psychopathology, Serious Mental Illness, and Suicidality: Large-Sample Cross-Sectional Study. Journal of Medical Internet Research. 2020;22(8):e17560. pmid:32780029
- View Article
- PubMed/NCBI
- Google Scholar
21. Wang BQ, Yao NQ, Zhou X, Liu J, Lv ZT. The association between attention deficit/hyperactivity disorder and Internet addiction: a systematic review and meta-analysis. BMC Psychiatry. 2017;17(1):260. pmid:28724403
- View Article
- PubMed/NCBI
- Google Scholar
22. Prizant-Passal S, Shechner T, Aderka IM. Social anxiety and Internet use–A meta-analysis: What do we know? What are we missing? Comput Human Behav. 2016;62:221–9.
- View Article
- Google Scholar
23. Ho RC, Zhang MWB, Tsang TY, Toh AH, Pan F, Lu Y, et al. The association between internet addiction and psychiatric co-morbidity: a meta-analysis. BMC Psychiatry. 2014;14(1):1–10. pmid:24947851
- View Article
- PubMed/NCBI
- Google Scholar
24. Hagenaars JA, McCutcheon AL. Applied Latent Class Analysis. Cambridge: Cambridge University Press; 2002.
25. McCutcheon A. Latent class analysis. Newbury Park, CA: SAGE; 1987.
26. Muthén B. Should substance use disorders be considered as categorical or dimensional? Addict. 2006;101(s1):6–16. pmid:16930156
- View Article
- PubMed/NCBI
- Google Scholar
27. Kardefelt-Winther D. Commentary on: Are we overpathologizing everyday life? A tenable blueprint for behavioral addiction research Problems with atheoretical and confirmatory research approaches in the study of behavioral addictions. J Behav Addict. 2015;4(3):126–9. pmid:26551896
- View Article
- PubMed/NCBI
- Google Scholar
28. Demetrovics Z, Király O, Koronczai B, Griffiths MD, Nagygyörgy K, Elekes Z, et al. Psychometric properties of the Problematic Internet Use Questionnaire Short-Form (PIUQ-SF-6) in a nationally representative sample of adolescents. PLoS ONE. 2016;11(8):e0159409. pmid:27504915
- View Article
- PubMed/NCBI
- Google Scholar
29. 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. International Journal of Mental Health. 2020:1–18.
- View Article
- Google Scholar
30. Göktaş S, Aygar H, Zencirci SA, Önsüz MF, Alaiye M, Metintaş S. Problematic internet use questionairre-short form-6 (PIUQ-SF 6): a validity and reliability study in Turkey. International Journal of Research in Medical Sciences. 2018;6(7):2354–60.
- View Article
- Google Scholar
31. Lin 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. 2020:105961. pmid:31072647
- View Article
- PubMed/NCBI
- Google Scholar
32. Moudiab S, Spada MM. The relative contribution of motives and maladaptive cognitions to levels of Internet Gaming Disorder. Addict Behav Rep. 2019;9:100160. pmid:30705935
- View Article
- PubMed/NCBI
- Google Scholar
33. Zsila Á, Orosz G, Király O, Urbán R, Ujhelyi A, Jármi É, et al. Psychoactive substance use and problematic Internet use as predictors of bullying and cyberbullying victimization. Int J Ment Health Addict. 2018;16(2):466–79. pmid:29670501
- View Article
- PubMed/NCBI
- Google Scholar
34. Macur M, Király O, Maraz A, Nagygyörgy K, Demetrovics Z. Prevalence of problematic Internet use in Slovenia. Slovenian Journal of Public Health. 2016;55(3):1–10. pmid:27703540
- View Article
- PubMed/NCBI
- Google Scholar
35. Macur M, Pontes HM. Individual differences and the development of Internet addiction: A nationally representative study. In: Bozoglan B, editor. Psychological, Social, and Cultural Aspects of Internet Addiction. Advances in Human and Social Aspects of Technology. Hershey, PA: IGI Global; 2018. p. 221–35.
36. Kim JE, Kim J. International note: Teen users’ problematic online behavior: Using panel data from South Korea. Journal of Adolescence. 2015;40:48–53. pmid:25621406
- View Article
- PubMed/NCBI
- Google Scholar
37. R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2020.
38. Muthén LK, Muthén BO. Mplus user’s guide (eighth edition). Los Angeles, CA: Muthén & Muthén; 2018.
39. Yentes RD, Wilhelm F. careless: Procedures for computing indices of careless responding. R package version 1.1.3. 2018.
40. Morey RD, Rouder JN, Jamil T, Urbanek S, Forner K, Ly A. BayesFactor: Computation of Bayes Factors for Common Designs. Retrieved from: https://cran.r-project.org/web/packages/BayesFactor/index.html. 2018.
- View Article
- Google Scholar
41. Revelle W. psych: Procedures for Psychological, Psychometric, and Personality Research. Evanston, Illinois: Northwestern University. R package version 2.0.9, https://CRAN.R-project.org/package=psych; 2020.
42. Pasek J. weights: Weighting and Weighted Statistics. Retrieved from: https://cran.r-project.org/web/packages/weights/index.html. 2020.
- View Article
- Google Scholar
43. Lee MD, Wagenmakers EJ. Bayesian cognitive modeling. Cambridge: Cambridge University Press; 2014.
44. Cohen J. Statistical power analysis for the behavioral sciences (2nd edition). New Jersey: Erlbaum: Hillsdale; 1988.
45. Akaike H. Factor analysis and AIC. Psychometrika. 1987;52(3):317–32.
- View Article
- Google Scholar
46. Schwarz G. Estimating the Dimension of a Model. Annals of Statistics. 1978;6(2):461–4.
- View Article
- Google Scholar
47. Sclove SL. Application of model-selection criteria to some problems in multivariate analysis. Psychometrika. 1987;52(3):333–43.
- View Article
- Google Scholar
48. Tein JY, Coxe S, Cham H. Statistical Power to Detect the Correct Number of Classes in Latent Profile Analysis. Structural Equation Modeling: A Multidisciplinary Journal. 2013;20(4):640–57. pmid:24489457
- View Article
- PubMed/NCBI
- Google Scholar
49. Shannon CE. A mathematical theory of communication. The Bell System Technical Journal. 1948;27(3):379–423. pmid:30854411
- View Article
- PubMed/NCBI
- Google Scholar
50. Asparouhov T, Muthén B. Variable-specific entropy contribution. Technical appendix. Los Angeles: Muthén & Muthén. Retrieved from: https://www.statmodel.com/download/UnivariateEntropy.pdf; 2014.
51. Lo Y, Mendell NR, Rubin DB. Testing the number of components in a normal mixture. Biometrika. 2001;88(3):767–78.
- View Article
- Google Scholar
52. Nylund KL, Asparouhov T, Muthén BO. Deciding on the Number of Classes in Latent Class Analysis and Growth Mixture Modeling: A Monte Carlo Simulation Study. Structural Equation Modeling: A Multidisciplinary Journal. 2007;14(4):535–69.
- View Article
- Google Scholar
53. Nagin D. Group-Based Modeling of Development. Cambridge (MA): Harvard University Press; 2009 30 Jun. 2009.
54. Nylund-Gibson K, Choi AY. Ten frequently asked questions about latent class analysis. Translational Issues in Psychological Science. 2018;4(4):440–61.
- View Article
- Google Scholar
55. Pontes HM, Caplan SE, Griffiths MD. Psychometric validation of the Generalized Problematic Internet Use Scale 2 in a Portuguese sample. Comput Human Behav. 2016;63:823–33.
- View Article
- Google Scholar
56. Pontes HM, Király O, Demetrovics Z, Griffiths MD. The conceptualisation and measurement of DSM-5 Internet Gaming Disorder: The development of the IGD-20 Test. PLoS ONE. 2014;9(10):e110137. pmid:25313515
- View Article
- PubMed/NCBI
- Google Scholar
57. Lin CY, Ganji M, Pontes HM, Imani V, Broström A, Griffiths MD, et al. Psychometric evaluation of the Persian Internet Disorder Scale among adolescents. J Behav Addict. 2018;7(3):665–75. pmid:30264609.
- View Article
- PubMed/NCBI
- Google Scholar
58. Fuster H, Carbonell X, Pontes HM, Griffiths MD. Spanish validation of the Internet Gaming Disorder-20 (IGD-20) Test. Comput Human Behav. 2016;56:215–24.
- View Article
- Google Scholar
59. Pontes HM, Griffiths MD. The development and psychometric evaluation of the Internet Disorder Scale (IDS-15). Addict Behav. 2017;64:261–8. pmid:26410796
- View Article
- PubMed/NCBI
- Google Scholar
60. Burkauskas J, Király O, Demetrovics Z, Podlipskyte A, Steibliene V. Psychometric Properties of the Nine-Item Problematic Internet Use Questionnaire (PIUQ-9) in a Lithuanian Sample of Students. Front Psychiatr. 2020;11(1279). pmid:33262711
- View Article
- PubMed/NCBI
- Google Scholar
61. Kirwil L. Parental Mediation Of Children’s Internet Use In Different European Countries. Journal of Children and Media. 2009;3(4):394–409.
- View Article
- Google Scholar
62. Livingstone S, Helsper EJ. Parental Mediation of Children’s Internet Use. Journal of Broadcasting & Electronic Media. 2008;52(4):581–99.
- View Article
- Google Scholar
63. Nielsen P, Favez N, Liddle H, Rigter H. Linking parental mediation practices to adolescents’ problematic online screen use: A systematic literature review. J Behav Addict. 2019:1–15. pmid:31786936
- View Article
- PubMed/NCBI
- Google Scholar
64. Starcevic V. Problematic Internet use: a distinct disorder, a manifestation of an underlying psychopathology, or a troublesome behaviour? World Psychiatry. 2010;9(2):92–3. pmid:20671892
- View Article
- PubMed/NCBI
- Google Scholar
65. Griffiths MD. A ‘components’ model of addiction within a biopsychosocial framework. Journal of Substance Use. 2005;10(4):191–7.
- View Article
- Google Scholar
66. Wang W, Li D, Li X, Wang Y, Sun W, Zhao L, et al. Parent-adolescent relationship and adolescent internet addiction: A moderated mediation model. Addict Behav. 2018;84:171–7. pmid:29709874
- View Article
- PubMed/NCBI
- Google Scholar
67. Huang S, Hu Y, Ni Q, Qin Y, Lü W. Parent-children relationship and internet addiction of adolescents: The mediating role of self-concept. Current Psychology. 2019.
- View Article
- Google Scholar
68. King DL, Delfabbro PH. Features of parent-child relationships in adolescents with Internet Gaming Disorder. Int J Ment Health Addict. 2016:1–14.
- View Article
- Google Scholar
69. Ioannidis K, Treder MS, Chamberlain SR, Kiraly F, Redden SA, Stein DJ, et al. Problematic internet use as an age-related multifaceted problem: Evidence from a two-site survey. Addict Behav. 2018;81:157–66. pmid:29459201
- View Article
- PubMed/NCBI
- Google Scholar
70. Pettorruso M, Valle S, Cavic E, Martinotti G, di Giannantonio M, Grant JE. Problematic Internet use (PIU), personality profiles and emotion dysregulation in a cohort of young adults: trajectories from risky behaviors to addiction. Psychiat Res. 2020;289:113036. pmid:32450451
- View Article
- PubMed/NCBI
- Google Scholar
71. Nakayama H, Matsuzaki T, Mihara S, Kitayuguchi T, Higuchi S. Relationship between problematic gaming and age at the onset of habitual gaming. Pediatrics International. 2020;62(11):1275–81. pmid:32379947
- View Article
- PubMed/NCBI
- Google Scholar
72. Liese BS, Kim HS, Hodgins DC. Insecure Attachment and Addiction: Testing the Mediating Role of Emotion Dysregulation in Four Potentially Addictive Behaviors. Addict Behav. 2020;107(11):106432. pmid:32330652
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Statista. Worldwide digital population as of October 2020 2020 [updated November 23November 27, 2020]. Available from: https://www.statista.com/statistics/617136/digital-population-worldwide.

[ref2] 2. Tifferet S. Gender Differences in Social Support on Social Network Sites: A Meta-Analysis. Cyberpsychol Behav Soc Netw. 2020;23(4):199–209. pmid:32096662
View Article
PubMed/NCBI
Google Scholar

[3] View Article

[4] PubMed/NCBI

[5] Google Scholar

[ref3] 3. Cauberghe V, Van Wesenbeeck I, De Jans S, Hudders L, Ponnet K. How Adolescents Use Social Media to Cope with Feelings of Loneliness and Anxiety During COVID-19 Lockdown. Cyberpsychol Behav Soc Netw. 2020. pmid:33185488
View Article
PubMed/NCBI
Google Scholar

[7] View Article

[8] PubMed/NCBI

[9] Google Scholar

[ref4] 4. Mandryk RL, Frommel J, Armstrong A, Johnson D. How Passion for Playing World of Warcraft Predicts In-Game Social Capital, Loneliness, and Wellbeing. Front Psychol. 2020;11(2165). pmid:33071843
View Article
PubMed/NCBI
Google Scholar

[11] View Article

[12] PubMed/NCBI

[13] Google Scholar

[ref5] 5. Karahanna E, Xu SX, Xu Y, Zhang NA. The needs–affordances–features perspective for the use of social media. MIS Quarterly: Management Information Systems. 2018;42(3):737–56.
View Article
Google Scholar

[15] View Article

[16] Google Scholar

[ref6] 6. Pontes HM, Kuss DJ, Griffiths MD. Clinical psychology of Internet addiction: A review of its conceptualization, prevalence, neuronal processes, and implications for treatment. Neuroscience and Neuroeconomics. 2015;4:11–23.
View Article
Google Scholar

[18] View Article

[19] Google Scholar

[ref7] 7. Starcevic V, Aboujaoude E. Internet addiction: Reappraisal of an increasingly inadequate concept. CNS Spectrums. 2016;22(1):7–13. pmid:26831456
View Article
PubMed/NCBI
Google Scholar

[21] View Article

[22] PubMed/NCBI

[23] Google Scholar

[ref8] 8. Starcevic V. Is Internet addiction a useful concept? Australian and New Zealand Journal of Psychiatry. 2013;47(1):16–9. pmid:23293309
View Article
PubMed/NCBI
Google Scholar

[25] View Article

[26] PubMed/NCBI

[27] Google Scholar

[ref9] 9. Griffiths MD. Internet "addiction": an issue for clinical psychology? Clinical Psychology Forum. 1996;97:32–6.
View Article
Google Scholar

[29] View Article

[30] Google Scholar

[ref10] 10. Young KS. Psychology of computer use: XL. Addictive use of the Internet: A case that breaks the stereotype. Psychological Reports. 1996;79(3):899–902. pmid:8969098
View Article
PubMed/NCBI
Google Scholar

[32] View Article

[33] PubMed/NCBI

[34] Google Scholar

[ref11] 11. Young KS, Pistner M, O’Mara J, Buchanan J. Cyber disorders: The mental health concern for the new millennium. Cyberpsychol Behav. 1999;2(5):475–9. pmid:19178220
View Article
PubMed/NCBI
Google Scholar

[36] View Article

[37] PubMed/NCBI

[38] Google Scholar

[ref12] 12. World Health Organization. International statistical classification of diseases and related health problems (11th ed.).2019.

[ref13] 13. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders (5th ed.)2013.

[ref14] 14. Pan YC, Chiu YC, Lin YH. Systematic Review and Meta-Analysis of Epidemiology of Internet Addiction. Neuroscience & Biobehavioral Reviews. 2020;118:612–22. pmid:32853626
View Article
PubMed/NCBI
Google Scholar

[42] View Article

[43] PubMed/NCBI

[44] Google Scholar

[ref15] 15. Su W, Han X, Jin C, Yan Y, Potenza MN. Are males more likely to be addicted to the internet than females? A meta-analysis involving 34 global jurisdictions. Comput Human Behav. 2019;99:86–100.
View Article
Google Scholar

[46] View Article

[47] Google Scholar

[ref16] 16. Ioannidis K, Hook R, Goudriaan AE, Vlies S, Fineberg NA, Grant JE, et al. Cognitive deficits in problematic internet use: meta-analysis of 40 studies. British Journal of Psychiatry. 2019;215(5):639–46. Epub 2019/02/20. pmid:30784392
View Article
PubMed/NCBI
Google Scholar

[49] View Article

[50] PubMed/NCBI

[51] Google Scholar

[ref17] 17. Alimoradi Z, Lin CY, Broström A, Bülow PH, Bajalan Z, Griffiths MD, et al. Internet addiction and sleep problems: A systematic review and meta-analysis. Sleep Medicine Reviews. 2019;47:51–61. pmid:31336284
View Article
PubMed/NCBI
Google Scholar

[53] View Article

[54] PubMed/NCBI

[55] Google Scholar

[ref18] 18. Lei H, Li S, Chiu MM, Lu M. Social support and Internet addiction among mainland Chinese teenagers and young adults: A meta-analysis. Comput Human Behav. 2018;85:200–9.
View Article
Google Scholar

[57] View Article

[58] Google Scholar

[ref19] 19. Fumero A, Marrero RJ, Voltes D, Peñate W. Personal and social factors involved in Internet addiction among adolescents: A meta-analysis. Comput Human Behav. 2018;86:387–400.
View Article
Google Scholar

[60] View Article

[61] Google Scholar

[ref20] 20. Guo W, Tao Y, Li X, Lin X, Meng Y, Yang X, et al. Associations of Internet Addiction Severity With Psychopathology, Serious Mental Illness, and Suicidality: Large-Sample Cross-Sectional Study. Journal of Medical Internet Research. 2020;22(8):e17560. pmid:32780029
View Article
PubMed/NCBI
Google Scholar

[63] View Article

[64] PubMed/NCBI

[65] Google Scholar

[ref21] 21. Wang BQ, Yao NQ, Zhou X, Liu J, Lv ZT. The association between attention deficit/hyperactivity disorder and Internet addiction: a systematic review and meta-analysis. BMC Psychiatry. 2017;17(1):260. pmid:28724403
View Article
PubMed/NCBI
Google Scholar

[67] View Article

[68] PubMed/NCBI

[69] Google Scholar

[ref22] 22. Prizant-Passal S, Shechner T, Aderka IM. Social anxiety and Internet use–A meta-analysis: What do we know? What are we missing? Comput Human Behav. 2016;62:221–9.
View Article
Google Scholar

[71] View Article

[72] Google Scholar

[ref23] 23. Ho RC, Zhang MWB, Tsang TY, Toh AH, Pan F, Lu Y, et al. The association between internet addiction and psychiatric co-morbidity: a meta-analysis. BMC Psychiatry. 2014;14(1):1–10. pmid:24947851
View Article
PubMed/NCBI
Google Scholar

[74] View Article

[75] PubMed/NCBI

[76] Google Scholar

[ref24] 24. Hagenaars JA, McCutcheon AL. Applied Latent Class Analysis. Cambridge: Cambridge University Press; 2002.

[ref25] 25. McCutcheon A. Latent class analysis. Newbury Park, CA: SAGE; 1987.

[ref26] 26. Muthén B. Should substance use disorders be considered as categorical or dimensional? Addict. 2006;101(s1):6–16. pmid:16930156
View Article
PubMed/NCBI
Google Scholar

[80] View Article

[81] PubMed/NCBI

[82] Google Scholar

[ref27] 27. Kardefelt-Winther D. Commentary on: Are we overpathologizing everyday life? A tenable blueprint for behavioral addiction research Problems with atheoretical and confirmatory research approaches in the study of behavioral addictions. J Behav Addict. 2015;4(3):126–9. pmid:26551896
View Article
PubMed/NCBI
Google Scholar

[84] View Article

[85] PubMed/NCBI

[86] Google Scholar

[ref28] 28. Demetrovics Z, Király O, Koronczai B, Griffiths MD, Nagygyörgy K, Elekes Z, et al. Psychometric properties of the Problematic Internet Use Questionnaire Short-Form (PIUQ-SF-6) in a nationally representative sample of adolescents. PLoS ONE. 2016;11(8):e0159409. pmid:27504915
View Article
PubMed/NCBI
Google Scholar

[88] View Article

[89] PubMed/NCBI

[90] Google Scholar

[ref29] 29. 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. International Journal of Mental Health. 2020:1–18.
View Article
Google Scholar

[92] View Article

[93] Google Scholar

[ref30] 30. Göktaş S, Aygar H, Zencirci SA, Önsüz MF, Alaiye M, Metintaş S. Problematic internet use questionairre-short form-6 (PIUQ-SF 6): a validity and reliability study in Turkey. International Journal of Research in Medical Sciences. 2018;6(7):2354–60.
View Article
Google Scholar

[95] View Article

[96] Google Scholar

[ref31] 31. Lin 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. 2020:105961. pmid:31072647
View Article
PubMed/NCBI
Google Scholar

[98] View Article

[99] PubMed/NCBI

[100] Google Scholar

[ref32] 32. Moudiab S, Spada MM. The relative contribution of motives and maladaptive cognitions to levels of Internet Gaming Disorder. Addict Behav Rep. 2019;9:100160. pmid:30705935
View Article
PubMed/NCBI
Google Scholar

[102] View Article

[103] PubMed/NCBI

[104] Google Scholar

[ref33] 33. Zsila Á, Orosz G, Király O, Urbán R, Ujhelyi A, Jármi É, et al. Psychoactive substance use and problematic Internet use as predictors of bullying and cyberbullying victimization. Int J Ment Health Addict. 2018;16(2):466–79. pmid:29670501
View Article
PubMed/NCBI
Google Scholar

[106] View Article

[107] PubMed/NCBI

[108] Google Scholar

[ref34] 34. Macur M, Király O, Maraz A, Nagygyörgy K, Demetrovics Z. Prevalence of problematic Internet use in Slovenia. Slovenian Journal of Public Health. 2016;55(3):1–10. pmid:27703540
View Article
PubMed/NCBI
Google Scholar

[110] View Article

[111] PubMed/NCBI

[112] Google Scholar

[ref35] 35. Macur M, Pontes HM. Individual differences and the development of Internet addiction: A nationally representative study. In: Bozoglan B, editor. Psychological, Social, and Cultural Aspects of Internet Addiction. Advances in Human and Social Aspects of Technology. Hershey, PA: IGI Global; 2018. p. 221–35.

[ref36] 36. Kim JE, Kim J. International note: Teen users’ problematic online behavior: Using panel data from South Korea. Journal of Adolescence. 2015;40:48–53. pmid:25621406
View Article
PubMed/NCBI
Google Scholar

[115] View Article

[116] PubMed/NCBI

[117] Google Scholar

[ref37] 37. R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2020.

[ref38] 38. Muthén LK, Muthén BO. Mplus user’s guide (eighth edition). Los Angeles, CA: Muthén & Muthén; 2018.

[ref39] 39. Yentes RD, Wilhelm F. careless: Procedures for computing indices of careless responding. R package version 1.1.3. 2018.

[ref40] 40. Morey RD, Rouder JN, Jamil T, Urbanek S, Forner K, Ly A. BayesFactor: Computation of Bayes Factors for Common Designs. Retrieved from: https://cran.r-project.org/web/packages/BayesFactor/index.html. 2018.
View Article
Google Scholar

[122] View Article

[123] Google Scholar

[ref41] 41. Revelle W. psych: Procedures for Psychological, Psychometric, and Personality Research. Evanston, Illinois: Northwestern University. R package version 2.0.9, https://CRAN.R-project.org/package=psych; 2020.

[ref42] 42. Pasek J. weights: Weighting and Weighted Statistics. Retrieved from: https://cran.r-project.org/web/packages/weights/index.html. 2020.
View Article
Google Scholar

[126] View Article

[127] Google Scholar

[ref43] 43. Lee MD, Wagenmakers EJ. Bayesian cognitive modeling. Cambridge: Cambridge University Press; 2014.

[ref44] 44. Cohen J. Statistical power analysis for the behavioral sciences (2nd edition). New Jersey: Erlbaum: Hillsdale; 1988.

[ref45] 45. Akaike H. Factor analysis and AIC. Psychometrika. 1987;52(3):317–32.
View Article
Google Scholar

[131] View Article

[132] Google Scholar

[ref46] 46. Schwarz G. Estimating the Dimension of a Model. Annals of Statistics. 1978;6(2):461–4.
View Article
Google Scholar

[134] View Article

[135] Google Scholar

[ref47] 47. Sclove SL. Application of model-selection criteria to some problems in multivariate analysis. Psychometrika. 1987;52(3):333–43.
View Article
Google Scholar

[137] View Article

[138] Google Scholar

[ref48] 48. Tein JY, Coxe S, Cham H. Statistical Power to Detect the Correct Number of Classes in Latent Profile Analysis. Structural Equation Modeling: A Multidisciplinary Journal. 2013;20(4):640–57. pmid:24489457
View Article
PubMed/NCBI
Google Scholar

[140] View Article

[141] PubMed/NCBI

[142] Google Scholar

[ref49] 49. Shannon CE. A mathematical theory of communication. The Bell System Technical Journal. 1948;27(3):379–423. pmid:30854411
View Article
PubMed/NCBI
Google Scholar

[144] View Article

[145] PubMed/NCBI

[146] Google Scholar

[ref50] 50. Asparouhov T, Muthén B. Variable-specific entropy contribution. Technical appendix. Los Angeles: Muthén & Muthén. Retrieved from: https://www.statmodel.com/download/UnivariateEntropy.pdf; 2014.

[ref51] 51. Lo Y, Mendell NR, Rubin DB. Testing the number of components in a normal mixture. Biometrika. 2001;88(3):767–78.
View Article
Google Scholar

[149] View Article

[150] Google Scholar

[ref52] 52. Nylund KL, Asparouhov T, Muthén BO. Deciding on the Number of Classes in Latent Class Analysis and Growth Mixture Modeling: A Monte Carlo Simulation Study. Structural Equation Modeling: A Multidisciplinary Journal. 2007;14(4):535–69.
View Article
Google Scholar

[152] View Article

[153] Google Scholar

[ref53] 53. Nagin D. Group-Based Modeling of Development. Cambridge (MA): Harvard University Press; 2009 30 Jun. 2009.

[ref54] 54. Nylund-Gibson K, Choi AY. Ten frequently asked questions about latent class analysis. Translational Issues in Psychological Science. 2018;4(4):440–61.
View Article
Google Scholar

[156] View Article

[157] Google Scholar

[ref55] 55. Pontes HM, Caplan SE, Griffiths MD. Psychometric validation of the Generalized Problematic Internet Use Scale 2 in a Portuguese sample. Comput Human Behav. 2016;63:823–33.
View Article
Google Scholar

[159] View Article

[160] Google Scholar

[ref56] 56. Pontes HM, Király O, Demetrovics Z, Griffiths MD. The conceptualisation and measurement of DSM-5 Internet Gaming Disorder: The development of the IGD-20 Test. PLoS ONE. 2014;9(10):e110137. pmid:25313515
View Article
PubMed/NCBI
Google Scholar

[162] View Article

[163] PubMed/NCBI

[164] Google Scholar

[ref57] 57. Lin CY, Ganji M, Pontes HM, Imani V, Broström A, Griffiths MD, et al. Psychometric evaluation of the Persian Internet Disorder Scale among adolescents. J Behav Addict. 2018;7(3):665–75. pmid:30264609.
View Article
PubMed/NCBI
Google Scholar

[166] View Article

[167] PubMed/NCBI

[168] Google Scholar

[ref58] 58. Fuster H, Carbonell X, Pontes HM, Griffiths MD. Spanish validation of the Internet Gaming Disorder-20 (IGD-20) Test. Comput Human Behav. 2016;56:215–24.
View Article
Google Scholar

[170] View Article

[171] Google Scholar

[ref59] 59. Pontes HM, Griffiths MD. The development and psychometric evaluation of the Internet Disorder Scale (IDS-15). Addict Behav. 2017;64:261–8. pmid:26410796
View Article
PubMed/NCBI
Google Scholar

[173] View Article

[174] PubMed/NCBI

[175] Google Scholar

[ref60] 60. Burkauskas J, Király O, Demetrovics Z, Podlipskyte A, Steibliene V. Psychometric Properties of the Nine-Item Problematic Internet Use Questionnaire (PIUQ-9) in a Lithuanian Sample of Students. Front Psychiatr. 2020;11(1279). pmid:33262711
View Article
PubMed/NCBI
Google Scholar

[177] View Article

[178] PubMed/NCBI

[179] Google Scholar

[ref61] 61. Kirwil L. Parental Mediation Of Children’s Internet Use In Different European Countries. Journal of Children and Media. 2009;3(4):394–409.
View Article
Google Scholar

[181] View Article

[182] Google Scholar

[ref62] 62. Livingstone S, Helsper EJ. Parental Mediation of Children’s Internet Use. Journal of Broadcasting & Electronic Media. 2008;52(4):581–99.
View Article
Google Scholar

[184] View Article

[185] Google Scholar

[ref63] 63. Nielsen P, Favez N, Liddle H, Rigter H. Linking parental mediation practices to adolescents’ problematic online screen use: A systematic literature review. J Behav Addict. 2019:1–15. pmid:31786936
View Article
PubMed/NCBI
Google Scholar

[187] View Article

[188] PubMed/NCBI

[189] Google Scholar

[ref64] 64. Starcevic V. Problematic Internet use: a distinct disorder, a manifestation of an underlying psychopathology, or a troublesome behaviour? World Psychiatry. 2010;9(2):92–3. pmid:20671892
View Article
PubMed/NCBI
Google Scholar

[191] View Article

[192] PubMed/NCBI

[193] Google Scholar

[ref65] 65. Griffiths MD. A ‘components’ model of addiction within a biopsychosocial framework. Journal of Substance Use. 2005;10(4):191–7.
View Article
Google Scholar

[195] View Article

[196] Google Scholar

[ref66] 66. Wang W, Li D, Li X, Wang Y, Sun W, Zhao L, et al. Parent-adolescent relationship and adolescent internet addiction: A moderated mediation model. Addict Behav. 2018;84:171–7. pmid:29709874
View Article
PubMed/NCBI
Google Scholar

[198] View Article

[199] PubMed/NCBI

[200] Google Scholar

[ref67] 67. Huang S, Hu Y, Ni Q, Qin Y, Lü W. Parent-children relationship and internet addiction of adolescents: The mediating role of self-concept. Current Psychology. 2019.
View Article
Google Scholar

[202] View Article

[203] Google Scholar

[ref68] 68. King DL, Delfabbro PH. Features of parent-child relationships in adolescents with Internet Gaming Disorder. Int J Ment Health Addict. 2016:1–14.
View Article
Google Scholar

[205] View Article

[206] Google Scholar

[ref69] 69. Ioannidis K, Treder MS, Chamberlain SR, Kiraly F, Redden SA, Stein DJ, et al. Problematic internet use as an age-related multifaceted problem: Evidence from a two-site survey. Addict Behav. 2018;81:157–66. pmid:29459201
View Article
PubMed/NCBI
Google Scholar

[208] View Article

[209] PubMed/NCBI

[210] Google Scholar

[ref70] 70. Pettorruso M, Valle S, Cavic E, Martinotti G, di Giannantonio M, Grant JE. Problematic Internet use (PIU), personality profiles and emotion dysregulation in a cohort of young adults: trajectories from risky behaviors to addiction. Psychiat Res. 2020;289:113036. pmid:32450451
View Article
PubMed/NCBI
Google Scholar

[212] View Article

[213] PubMed/NCBI

[214] Google Scholar

[ref71] 71. Nakayama H, Matsuzaki T, Mihara S, Kitayuguchi T, Higuchi S. Relationship between problematic gaming and age at the onset of habitual gaming. Pediatrics International. 2020;62(11):1275–81. pmid:32379947
View Article
PubMed/NCBI
Google Scholar

[216] View Article

[217] PubMed/NCBI

[218] Google Scholar

[ref72] 72. Liese BS, Kim HS, Hodgins DC. Insecure Attachment and Addiction: Testing the Mediating Role of Emotion Dysregulation in Four Potentially Addictive Behaviors. Addict Behav. 2020;107(11):106432. pmid:32330652
View Article
PubMed/NCBI
Google Scholar

[220] View Article

[221] PubMed/NCBI

[222] Google Scholar

Figures

Abstract

Objective

Methods

Results

Conclusions

Introduction

The present study

Material and methods

Participants

Measures

Sociodemographic and internet use data.

Problematic internet use.

Subjective well-being.

Parent-child relationship.

Self-control.

Statistical analysis

Results

Descriptive statistics

LPA

Determining the number of latent classes.

Latent class characterization and interpretation.

Cross-class comparisons.

Discussion

Conclusion

References