Skip to Content
MDPIMDPI

Advanced

AdolescentsAdolescents
PDF
  • Article
  • Open Access

11 May 2026

14 Pages

Parental Reports of Adolescent Problematic Internet Use: Associations with Emotional and Behavioral Problems and Family Functioning in Cyprus

,
,
and
1
School of Medicine, National and Kapodistrian University of Athens, 115 27 Athens, Greece
2
Department of Child and Adolescent Psychiatry, School of Medicine, National and Kapodistrian University of Athens, 115 27 Athens, Greece
3
Department of Psychiatry, Faculty of Medicine, University of Thessaly, 377 87 Larissa, Greece
*
Author to whom correspondence should be addressed.
Adolescents2026, 6(3), 41;https://doi.org/10.3390/adolescents6030041
Version Notes
Review Reports
Order Reprints

Abstract

Excessive internet use among adolescents is a growing public health concern, yet data from Cyprus are scarce. This cross-sectional study used parent reports for 248 adolescents (12–17 years; 54.4% boys) to describe problematic internet use and examine associations with emotional–behavioral difficulties and family functioning. Based on published Parent–Child Internet Addiction Test (PCIAT) cut-offs, used descriptively rather than as clinically validated Cypriot thresholds, 74.2% of adolescents in this parent-reported convenience sample fell within the broader problematic internet use range. This sample-specific proportion comprised 38.3% in the mild, 33.9% in the moderate, and 2.0% in the severe PCIAT category. Adolescents using the internet >3 h/day (n = 112) had significantly higher PCIAT scores than those using it <3 h/day (n = 136; U = 3100.00, Z = −4.72, p < 0.001). PCIAT scores correlated with conduct problems (r = 0.449), hyperactivity/inattention (r = 0.441), and total difficulties (r = 0.475), and inversely with prosocial behavior (r = −0.471; p < 0.001). Family functioning worsened across PCIAT severity categories (GF-12 means: 18.83 in the PCIAT 20–30 range vs. 29.00 in the severe range; H(3) = 28.30, p < 0.001). In multivariable regression, daily use (B = 8.02), poorer family functioning (B = 0.42), conduct problems (B = 5.28), and hyperactivity/inattention (B = 6.81) independently predicted higher PCIAT scores (R2 = 0.24), whereas emotional symptoms were not significant (p = 0.637). Findings from this convenience sample highlight the potential importance of prevention efforts targeting screen-time regulation and family relationships; longitudinal, representative, multi-informant studies are warranted.

1. Introduction

In recent decades, digital technologies have dramatically reshaped everyday life, creating unprecedented opportunities for communication, education, and entertainment [1,2]. The internet, once a specialized tool, has evolved into an essential resource that permeates nearly every aspect of modern society. Adolescents, who are early adopters of digital media, have become particularly immersed in the online world [3,4,5]. While the benefits of digital connectivity are numerous, excessive and unregulated internet use has given rise to a growing public health concern often described as problematic internet use, internet addiction, or pathological internet use [6,7]. This pattern is characterized by difficulties in controlling online behavior, which may be associated with impairments in social, academic, and emotional domains [8,9,10].
It is important to note that the terms “internet addiction,” “pathological internet use,” and “problematic internet use” have been used variably in the literature [11,12]. Because “internet addiction” may imply a clinically defined disorder, whereas the present study is based on parent-reported questionnaire scores rather than clinical diagnosis, we use “problematic internet use” as the main umbrella term throughout the manuscript [13]. The term “internet addiction” is retained where necessary when referring to the Parent–Child Internet Addiction Test (PCIAT), its original terminology, and the PCIAT severity categories derived from published cut-offs [14]. This approach allows comparison with earlier foundational studies while avoiding overdiagnostic interpretation of the findings.
The concept of problematic internet use, historically often described as internet addiction, emerged in the mid-1990s and has since evolved through various theoretical frameworks [15,16]. Early research primarily focused on establishing diagnostic criteria and exploring neurobiological and cognitive-behavioral correlates of excessive internet use, drawing parallels with other addictive behaviors by emphasizing mechanisms such as reward sensitivity and impaired self-regulation [17]. Although a full biopsychosocial account of problematic internet use would include biological, psychological, family, peer, school, and broader sociocultural factors [15,16], the present study does not aim to test such a comprehensive model. Rather, it is informed by a developmental and psychosocial perspective and focuses specifically on parent-reported emotional–behavioral difficulties and family functioning as correlates of problematic internet use in adolescents.
Studies have demonstrated that problematic internet use is often associated with psychiatric conditions such as depression, anxiety, and attention deficit/hyperactivity disorder, as well as with externalizing behaviors like impulsivity and conduct problems [18,19,20]. More recently, attention has shifted to the role of the family environment in either exacerbating or mitigating the risk of problematic internet use [21,22,23,24]. Family dynamics—including the quality of communication, emotional support, and conflict resolution—appear to be critical determinants in how adolescents manage the challenges of modern digital life [25,26]. Research conducted in various cultural settings has consistently shown that dysfunctional family environments are associated with higher levels of problematic internet use among adolescents; one possible interpretation is that lower perceived support and poorer family communication may co-occur with greater reliance on online activities for social connection or emotional regulation [27,28,29,30,31].
Despite a robust body of international literature, there remains a scarcity of research examining problematic internet use among adolescents within Cyprus [32]. The sociocultural context of Cyprus, with its distinct family structures, parenting styles, and educational practices, may influence both the distribution and the manifestation of problematic internet use in unique ways [33]. It is crucial to explore these relationships locally, as understanding the specific interplay between individual predispositions, emotional–behavioral functioning, and family dynamics in Cyprus will provide valuable insights for tailoring prevention and intervention strategies that are culturally relevant.
The current study aims to address these gaps by describing the distribution of parent-reported problematic internet use severity in a convenience sample of Cypriot adolescents and by examining the associations between PCIAT scores, emotional–behavioral problems, and family functioning. Accordingly, the study focuses on selected psychosocial correlates rather than attempting to operationalize the full range of biological, peer, school, or broader sociocultural factors that may also be relevant to problematic internet use. In addition to exploring how factors such as excessive daily internet use, poor family functioning, and externalizing behaviors are associated with problematic internet use, this study seeks to assess whether internalizing symptoms (e.g., depression and anxiety) are also significant predictors when evaluated alongside behavioral issues.
Based on the extant literature and preliminary observations, we hypothesize that: first, a substantial proportion of adolescents in this parent-reported convenience sample will fall within the problematic internet use range, with some reaching moderate or severe PCIAT categories; second, adolescents who spend more than three hours per day online will demonstrate higher PCIAT scores compared to those with lower daily usage; third, poorer family functioning will be significantly associated with greater problematic internet use severity; and fourth, externalizing behavioral problems—specifically, conduct problems and hyperactivity/inattention—will be positively correlated with problematic internet use, while internalizing symptoms may not emerge as significant predictors once other variables are controlled. These hypotheses will be tested using a cross-sectional design, with data collected from parents of adolescents in Cyprus, thereby providing preliminary, sample-specific insights that may inform future culturally appropriate intervention and prevention research.

2. Materials and Methods

2.1. Study Design and Participants

This study was designed as a cross-sectional investigation aimed at examining the distribution and correlates of parent-reported problematic internet use among adolescents in the Republic of Cyprus. Data were collected from parents or legal guardians of adolescents aged between 12 and 17 years. A convenience sample was obtained, including families from different regions of the Republic of Cyprus. Because recruitment and data collection were conducted online, participation was necessarily limited to parents or legal guardians with internet access and sufficient digital literacy to complete an electronic survey. The inclusion criteria required that participants be the parent or legal guardian of at least one child within the specified age range, have sufficient proficiency in the Greek language, and be permanent residents of Cyprus. In cases where parents had more than one eligible child, they were instructed to complete the survey with respect to the oldest child in the target age range. This strategy was employed to maintain consistency in the selection process and to minimize within-family variability in reporting. Because the study used an online convenience sample, it was not designed to be representative of all adolescents or families in Cyprus, particularly families with limited internet access, lower digital literacy, or lower educational attainment. Therefore, the observed PCIAT category distribution should not be interpreted as a population prevalence estimate.

2.2. Recruitment and Ethical Approval

Recruitment was conducted through various online platforms including social media channels (such as Facebook), email distribution lists, and messaging applications. This multi-channel approach was chosen to ensure wide reach and to facilitate participation among parents from diverse geographic and socioeconomic backgrounds; however, as participation was voluntary and recruitment was conducted online, the possibility of self-selection bias cannot be excluded. Prior to data collection, potential participants received an electronic invitation that included a brief description of the study’s objectives, procedures, and the voluntary nature of participation. The study protocol, along with the recruitment procedures and consent materials, was reviewed and approved by the National Bioethics Committee of Cyprus (Approval No. ΕΕΒΚ ΕΠ 2022.01.143). All participants provided informed consent electronically before proceeding to the survey, ensuring that ethical standards regarding confidentiality, voluntariness, and the right to withdraw at any time were strictly upheld.

2.3. Measures

The survey instrument was developed using Google Forms and comprised several sections to collect a comprehensive dataset. The first section gathered demographic information about the parent (age, gender, marital status, education, economic status) and the adolescent (age, gender, and number of children in the family). Additional items probed parental perceptions of the adolescent’s online behavior and academic performance, as well as details regarding the adolescent’s recent engagement with mental health services. Recent mental health service use was assessed with a single yes/no item asking whether the adolescent had visited a mental health specialist during the previous six months. This item was included to describe the help-seeking profile of the sample and was not intended to indicate psychiatric diagnosis, symptom severity, or treatment status. Parents also reported the adolescent’s average daily internet use. In the survey, daily use was recorded categorically as less than three hours per day versus more than three hours per day. This cut-off was chosen a priori as a pragmatic threshold to distinguish lower from higher daily internet exposure while preserving sufficient group sizes for comparison. Because the item was collected categorically, daily internet use could not be re-analysed as a continuous variable.
The second section of the survey included the Parent–Child Internet Addiction Test (PCIAT), a questionnaire designed for parental assessment of adolescents’ online behavior [14,34]. This instrument consists of 20 items, each rated on a five-point Likert scale ranging from 1 (“rarely”) to 5 (“always”). Total scores can range from 20 to 100. In line with the published PCIAT/IAT-F scoring approach, scores were descriptively grouped into four severity categories: non-addicted range (20–30), mild range (31–49), moderate range (50–79), and severe range (80–100) [14,34]. The questionnaire was translated into Greek using forward–backward translation procedures. In the present study, the PCIAT demonstrated excellent internal consistency (Cronbach’s α = 0.96). However, the Greek version has not been formally validated in Cypriot adolescents; therefore, internal consistency should not be interpreted as evidence of content validity, construct validity, measurement invariance, or clinical validity of the cut-off scores in this population.
The third section employed the parent-report version of the Strengths and Difficulties Questionnaire (SDQ) to assess the emotional and behavioral functioning of the adolescent [35,36]. The Greek version of the SDQ has previously demonstrated acceptable psychometric properties in Greek adolescents and their parents [36]. In the present sample, internal consistency varied across SDQ scales, as is often observed for brief multi-item screening subscales. Cronbach’s α was 0.72 for emotional symptoms, 0.47 for conduct problems, 0.66 for hyperactivity/inattention, 0.46 for peer problems, 0.69 for prosocial behavior, and 0.80 for total difficulties. Thus, while the total difficulties score showed good internal consistency, some brief SDQ subscales, particularly conduct problems and peer problems, showed low reliability and should be interpreted cautiously.
The final standardized instrument was the General Functioning Scale (GF-12) of the McMaster Family Assessment Device (FAD) [37,38]. The GF-12 consists of 12 items that evaluate overall family functioning, including communication, problem-solving, and emotional support. Items are rated on a 4-point Likert scale, and higher total scores reflect poorer family functioning. Furthermore, the GF-12 exhibited excellent internal consistency in our sample (Cronbach’s α = 0.87), supporting its reliability in evaluating overall family functioning among Cypriot families.

2.4. Data Collection Procedures

Data collection was carried out online over a three-month period from June to August 2022. The survey was hosted on Google Forms (Google LLC, Mountain View, CA, USA) and was accessible via a provided link in the electronic invitation. Participants were able to complete the survey on a variety of devices including smartphones, tablets, and desktop computers. The survey was structured in a user-friendly format and was estimated to take approximately 20 min to complete. Once completed, responses were automatically recorded in the Google Forms response database and were later exported to spreadsheet format for analysis. To ensure the confidentiality of the participants, no personally identifiable information was collected, and all responses were stored in a password-protected account accessible only to the research team.

2.5. Statistical Analysis

Data analysis was performed using SPSS, version 29.0 (IBM Corp., Armonk, NY, USA). Descriptive statistics, including means, standard deviations, medians, and interquartile ranges, were computed to summarize the continuous variables such as age and test scores. Frequencies and percentages were calculated for categorical variables, including gender, marital status, and descriptively defined PCIAT severity categories. The normality of continuous variables was assessed using the Kolmogorov–Smirnov test. For comparisons of categorical data, Pearson’s chi-square tests were conducted, with Fisher’s exact test applied when the expected frequencies were too low to meet chi-square assumptions. Comparisons of quantitative data between two groups were made using the Mann–Whitney U test, while the Kruskal–Wallis test was used when more than two groups were compared. When Kruskal–Wallis tests were statistically significant, Dunn–Bonferroni post hoc pairwise comparisons were performed. Spearman’s rank correlation coefficients were calculated to examine the relationships between the PCIAT scores and the subscale scores of the SDQ. To identify independent predictors of PCIAT scores, a stepwise multiple linear regression analysis was performed with the PCIAT score serving as the dependent variable. Independent variables included adolescent gender, age, daily internet use, family functioning (GF-12 score), and the subscale scores from the SDQ. Recent mental health service use was not included in the regression model because it was assessed only as a broad, binary help-seeking indicator, was reported by a relatively small subgroup of the sample, and did not provide information about diagnosis, reason for consultation, treatment type, or clinical severity. Daily internet use was entered as a dichotomous variable because the original survey recorded it categorically as less than three hours versus more than three hours per day. The regression model generated unstandardized regression coefficients (B), standard errors, and significance levels (p-values) to indicate the relative contribution of each variable to the overall variance in PCIAT scores. All statistical tests were two-tailed, with the level of significance set at p < 0.05.

3. Results

3.1. Demographic Characteristics

The final sample comprised 248 parents (M = 45.4 years, SD = 5.5), of whom 207 (83.5%) were female and 41 (16.5%) were male. Parents reported on one adolescent (M = 14.6 years, SD = 1.6), with 135 (54.4%) being boys and 113 (45.6%) girls. Most parents were married (n = 215, 86.7%) and had attained tertiary education (n = 208, 83.9%). This indicates that the sample was highly educated and may not reflect the educational distribution of the broader parent population in Cyprus. In addition, 23 parents (9.3%) reported a low, 183 (73.8%) a medium, and 42 (16.9%) a high economic status, and 148 families (59.7%) had one to two children. Furthermore, 33 adolescents (13.3%) had visited a mental health specialist in the past six months (see Table 1). This variable was used to describe recent help-seeking in the sample and was not treated as a diagnostic indicator.
Table 1. Demographic characteristics of parents and adolescents (N = 248).

3.2. Online Activities

Over the past three months, parents reported that 76.2% of adolescents watched online content (e.g., YouTube, Netflix), 68.5% used social networking sites (e.g., Facebook, Instagram), 60.1% engaged in online gaming, 0% participated in online gambling, 40.7% searched for general information, and 44.4% searched for school assignments (see Table 2).
Table 2. Parent-reported online activities of adolescents.

3.3. Overall Descriptive Statistics of Main Study Variables

Overall descriptive statistics for the main study variables are presented in Table 3. The mean PCIAT total score was 44.07, indicating that the average score in this parent-reported convenience sample fell within the mild PCIAT range according to the published cut-offs used descriptively. Mean SDQ scores indicated relatively low levels of emotional–behavioral difficulties overall, while the mean GF-12 score was 21.40, with higher GF-12 scores indicating poorer family functioning.
Table 3. Descriptive statistics for PCIAT, SDQ, and GF-12 scores in the total sample.

3.4. Problematic Internet Use Severity

Based on published PCIAT cut-offs used descriptively, 25.8% of adolescents were classified in the non-addicted range (scores 20–30), 38.3% in the mild range (scores 31–49), 33.9% in the moderate range (scores 50–79), and 2.0% in the severe range (scores 80–100). Overall, within the present convenience sample, 74.2% of adolescents fell within the mild, moderate, or severe PCIAT range, indicating parent-reported problematic internet use according to the applied descriptive cut-offs. A Mann–Whitney U test comparing PCIAT scores between adolescents using the internet for more than three hours per day (n = 112) and those using it for less than three hours (n = 136) yielded U = 3100.00, Z = −4.72, p < 0.001, indicating that higher daily internet use was significantly associated with elevated PCIAT scores (see Table 4).
Table 4. Parent-reported PCIAT severity categories in the study sample.

3.5. Emotional and Behavioral Functioning

Parent reports on the Strengths and Difficulties Questionnaire (SDQ) showed that girls scored higher on emotional symptoms (M = 2.77, SD = 2.3) than boys (M = 1.74, SD = 1.8). A Mann–Whitney U test for emotional symptoms revealed U = 4050.00, Z = −3.50, p = 0.001. In contrast, boys exhibited higher hyperactivity/inattention scores (M = 3.01, SD = 2.2) than girls (M = 2.38, SD = 1.9), with U = 4200.00, Z = −2.00, p = 0.047. No statistically significant gender differences emerged for conduct problems (U = 5000.00, Z = −0.37, p = 0.717), peer problems (U = 5100.00, Z = −0.80, p = 0.421), prosocial behavior (U = 4950.00, Z = −0.76, p = 0.448), or total difficulties (U = 5025.00, Z = −0.65, p = 0.516) (see Table 5).
Table 5. SDQ subscale scores by adolescent gender (parent-reported).

3.6. Correlational Analyses

Spearman’s rank correlation coefficients indicated that PCIAT scores were significantly positively correlated with hyperactivity/inattention, r(246) = 0.441, p < 0.001; conduct problems, r(246) = 0.449, p < 0.001; peer problems, r(246) = 0.318, p < 0.001; total difficulties, r(246) = 0.475, p < 0.001; and poorer family functioning as measured by the GF-12, r(246) = 0.354, p < 0.001. In contrast, prosocial behavior was significantly negatively correlated with PCIAT scores, r(246) = −0.471, p < 0.001, and the correlation with emotional symptoms was small but statistically significant, r(246) = 0.112, p = 0.045 (see Table 6).
Table 6. Spearman’s correlations between PCIAT scores and emotional–behavioral/family functioning variables.

3.7. Family Functioning

In addition to the overall positive correlation between PCIAT and GF-12 scores reported above, the association between family functioning (GF-12 scores) and PCIAT severity categories was examined using a Kruskal–Wallis test. Adolescents in the PCIAT 20–30 range had a mean GF-12 score of 18.83 (SD = 6.57; median = 18, IQR: 13–23), those in the mild PCIAT range had a mean score of 21.26 (SD = 5.15; median = 21, IQR: 17–24), those in the moderate range had a mean score of 23.05 (SD = 6.07; median = 23, IQR: 19.75–26), and those in the severe range had a mean score of 29.00 (SD = 6.86; median = 30, IQR: 28–33). The Kruskal–Wallis test yielded H(3) = 28.30, p < 0.001, indicating statistically significant differences among the groups. Dunn–Bonferroni post hoc comparisons showed that the PCIAT 20–30 range differed significantly from the mild range (p = 0.011), moderate range (p < 0.001), and severe range (p = 0.005). Pairwise differences among the mild, moderate, and severe PCIAT ranges were not statistically significant after Bonferroni correction (all p ≥ 0.132).

3.8. Regression Analysis

A stepwise multiple linear regression was conducted with PCIAT score as the dependent variable and daily internet use (dichotomized as >3 h vs. <3 h), GF-12 score, and SDQ subscale scores (conduct problems, hyperactivity/inattention, and emotional symptoms) as predictors. The overall regression model was statistically significant, F(5, 242) = 15.67, p < 0.001, explaining approximately 24% of the variance in PCIAT scores (R2 = 0.24), indicating a modest proportion of explained variance. Daily internet use emerged as the strongest predictor (B = 8.017, SE = 1.803, t(242) = 4.45, p < 0.001), followed by family functioning (B = 0.416, SE = 0.154, t(242) = 2.70, p = 0.007), conduct problems (B = 5.280, SE = 2.115, t(242) = 2.50, p = 0.013), and hyperactivity/inattention (B = 6.811, SE = 2.904, t(242) = 2.34, p = 0.020). Emotional symptoms were not independently associated with PCIAT scores in the multivariable model (B = 1.163, SE = 2.459, t(242) = 0.47, p = 0.637) (see Table 7).
Table 7. Multiple linear regression predicting PCIAT scores (N = 248).

4. Discussion

This study provides preliminary insight into the distribution of parent-reported problematic internet use severity in a convenience sample of adolescents in Cyprus and elucidates its associations with emotional–behavioral problems and family functioning. In our sample, approximately 74% of adolescents fell within the broader problematic-use range according to published PCIAT cut-offs, while 2% were classified in the severe PCIAT category. This proportion should be interpreted with particular caution. First, the broader problematic-use range combines mild, moderate, and severe PCIAT categories and therefore should not be equated with clinically diagnosed internet addiction. Second, given the online convenience sampling strategy, the proportion may have been influenced by self-selection of digitally engaged parents or parents especially concerned about their adolescent’s internet use. Third, because the Greek PCIAT has not been formally validated in Cypriot adolescents, the applicability of the published cut-offs to this cultural and linguistic context remains uncertain. Therefore, the observed distribution should be understood as a sample-specific finding rather than as a representative prevalence estimate for Cyprus. Excessive daily internet use, specifically more than three hours per day, was the strongest statistical correlate of higher PCIAT scores in the regression model. In addition, poorer family functioning was independently associated with elevated PCIAT scores. This was also supported by the bivariate correlation between PCIAT and GF-12 scores, indicating that poorer family functioning was associated with higher parent-reported problematic internet use. Post hoc comparisons further indicated that the clearest differences in family functioning were observed between adolescents in the PCIAT 20–30 range and those in the mild, moderate, or severe PCIAT ranges, whereas differences among the mild, moderate, and severe ranges were not statistically significant after correction. This is consistent with previous research indicating that supportive family environments are associated with lower levels of problematic internet use, whereas poorer family functioning is associated with higher levels of problematic internet use. In addition, Cypriot family, educational, and social norms may shape both adolescents’ online behavior and parents’ perceptions of such behavior; however, these contextual factors were not directly measured in the present study.
The descriptive pattern of online activities provides useful context for interpreting the PCIAT findings. Parents most frequently reported adolescents’ use of online content platforms, social networking sites, and online gaming, whereas information-seeking and school-related use were less frequently reported. These patterns suggest that much of the reported internet use may have involved recreational, social, and entertainment-oriented activities rather than purely educational use. Such activities may be particularly relevant to problematic patterns of use because they are often highly engaging, socially reinforcing, and difficult for adolescents to self-regulate. However, because the present study did not examine specific online activities as independent predictors of PCIAT scores, these observations should be interpreted descriptively. Future studies should investigate whether different types of online activity, such as gaming, social networking, streaming, or school-related use, show distinct associations with problematic internet use and emotional–behavioral functioning.
The results further suggest that externalizing behaviors, particularly conduct problems and hyperactivity/inattention, were more strongly associated with PCIAT scores than emotional symptoms. Although many studies have found links between internalizing symptoms (e.g., anxiety, depression) and problematic internet use [18,19,20], in the present study emotional symptoms showed only a small bivariate association with PCIAT scores and did not emerge as an independent predictor in the regression model. This finding should be interpreted cautiously. The loss of statistical significance in the multivariable model may reflect shared variance with externalizing symptoms and family functioning, possible suppression or overlap among predictors, or the parent-report format, since parents may be less able to detect adolescents’ internal distress than observable behavioral difficulties. Although the emotional symptoms subscale showed acceptable internal consistency in the present sample, it remains a brief parent-reported screening scale and may not fully capture adolescents’ internalizing distress. Therefore, the non-significant regression coefficient for emotional symptoms should not be interpreted as evidence that internalizing symptoms are unrelated to problematic internet use. Future studies should incorporate adolescent self-reports, clinician assessments, and more comprehensive measures of anxiety and depression.
The significant association between poorer family functioning and problematic internet use supports the relevance of the family environment in understanding adolescent online behavior. Families characterized by lower cohesion, poorer communication, and greater conflict have been found in previous studies to show higher levels of adolescent problematic internet use. These findings are in line with earlier studies showing that supportive home environments are associated with lower levels of several maladaptive behaviors [21,22,23,24,25,26]. In contrast, dysfunctional family dynamics may be associated with adolescents’ greater reliance on online activities, which in turn may relate to increased screen time and problematic patterns of use [27,28,29,30,31].
The multiple regression analysis indicated that daily internet use, family functioning, and externalizing behaviors were independently associated with PCIAT scores. However, the model explained a modest proportion of variance (R2 = 0.24), indicating that a substantial part of the variability in problematic internet use was not captured by the variables included in this study. Important unmeasured factors may include peer influence, school climate, parental screen use and digital rules, adolescent motives for internet use, sleep problems, academic stress, and broader sociocultural influences. This pattern of associations suggests that prevention and intervention approaches may benefit from considering not only screen-time regulation but also family dynamics and behavioral problems, while future longitudinal studies are needed to determine temporal and causal pathways.
There are several limitations to consider. The cross-sectional design does not permit the establishment of causality or directionality among problematic internet use, emotional–behavioral difficulties, and family functioning. Another limitation of this study is its exclusive reliance on parent reports. Parents may not be fully aware of their adolescents’ internal experiences—especially internalizing symptoms such as anxiety and depression—which could lead to underreporting of these issues. Consequently, the findings might not fully capture the internal distress experienced by the adolescents. Moreover, the small bivariate association between emotional symptoms and PCIAT scores, combined with the non-significant regression coefficient, may reflect shared variance with other predictors, possible suppression effects, and the limitations of brief parent-reported screening scales rather than the absence of a meaningful relationship. In addition, the relatively low internal consistency of some SDQ subscales, particularly conduct problems and peer problems, may have attenuated some observed associations and should be considered when interpreting the SDQ-related findings. Although recent mental health specialist visits were recorded, this variable was limited to a single binary indicator of help-seeking and did not capture psychiatric diagnosis, clinical severity, treatment type, or reasons for consultation; therefore, it was not included as an explanatory variable in the main analyses. Future studies should incorporate multiple informants, including adolescent self-reports and clinician assessments, and should collect more detailed clinical and service-use information to provide a more comprehensive evaluation of problematic internet use and its correlates.
Additionally, the online convenience sampling method may have introduced important participation and self-selection bias. Because the survey was distributed and completed online, parents without regular internet access or with lower digital literacy were less likely to be reached or able to participate. The sample was also highly educated, with 83.9% of participating parents reporting tertiary education. This educational skew may limit representativeness, because highly educated parents may differ from the broader parent population in Cyprus in digital literacy, awareness of problematic internet use, parenting practices, access to resources, willingness to participate in research, and perceptions or reporting of adolescent online behavior. Moreover, parents who were especially concerned about their adolescent’s internet use may have been more motivated to complete the survey. Consequently, both the observed PCIAT category distribution and the associations with family functioning and emotional–behavioral difficulties may overestimate, underestimate, or otherwise differ from patterns in the broader population of Cypriot adolescents. These findings should therefore be interpreted as sample-specific and should not be generalized to all Cypriot adolescents or families, particularly those with lower educational attainment, lower digital engagement, or limited access to online survey participation.
In addition, although the PCIAT demonstrated excellent internal consistency in our sample, the Greek version has not been formally validated in Cypriot adolescents. Cronbach’s α provides evidence of internal consistency only and does not establish content validity, construct validity, measurement invariance, or the appropriateness of the published cut-off scores for this population. Therefore, the PCIAT severity categories should be interpreted descriptively rather than as clinically validated categories for Cypriot adolescents. This limitation is especially important when interpreting the high proportion of adolescents classified in the mild-or-above PCIAT range, because cut-off performance may vary across cultural, linguistic, and social contexts.
Another limitation concerns the measurement of daily internet use. Daily use was assessed using a dichotomous item (<3 h/day vs. >3 h/day), which reduced variability and did not allow examination of dose–response associations across the full range of use. Although this cut-off provided a pragmatic contrast between lower and higher daily use, future studies should assess internet use as a continuous or more finely categorized variable, ideally distinguishing educational, social, gaming, entertainment, and other online activities.
Lastly, while the regression model was statistically significant, it explained only a modest proportion of the variance in PCIAT scores (R2 = 0.24). This indicates that most of the variability in parent-reported problematic internet use was not explained by the predictors included in the model. The study did not assess several potentially important factors, including neurocognitive characteristics, adolescent motives for internet use, sleep problems, academic stress, parental screen use, parental monitoring and digital rules, peer influence, school climate, and broader sociocultural dynamics. In addition, although online activities were described, the study did not examine whether specific types of internet use, such as gaming, social networking, streaming, or school-related use, were differentially associated with PCIAT scores. These unmeasured or unanalysed variables may have contributed substantially to problematic internet use and should be investigated in future longitudinal and multi-informant research.
Despite these limitations, the study offers several important implications for prevention and intervention. Given the high proportion of adolescents in our sample who fell within the problematic-use range, together with the observed associations with daily screen time and family dysfunction, educational institutions, healthcare providers, and policymakers may consider collaborative prevention efforts that promote healthy internet use. Family-based interventions that focus on communication, conflict reduction, and clear, supportive guidelines for internet use may be clinically relevant, although their effectiveness in this population should be tested in future intervention studies. It is crucial that such interventions avoid overly punitive measures; instead, they may help parents set realistic and flexible limits that support adolescents’ development of self-regulation skills.
The current findings also suggest that clinicians treating adolescents for problematic internet use should consider evaluating family dynamics as part of the assessment process. Programs that incorporate both individual and family therapy components could be adapted for use in Cyprus. The data are consistent with the view that clinical assessment of problematic internet use may benefit from a multi-pronged approach that considers both behavioral symptoms and the family context.
Future research should build on the present findings by employing longitudinal designs to explore the temporal relationship between family functioning and problematic internet use. Multi-informant approaches, including both parent and adolescent self-reports, will be critical in obtaining a more comprehensive picture of problematic internet use and its correlates. In addition, intervention studies that rigorously test the efficacy of combined family and individual therapies will provide valuable guidance for clinical practice. Finally, further investigation into the influence of external factors such as peer relationships, school environment, and recent societal changes (e.g., the impact of the COVID-19 pandemic) is needed to fully understand the complexity of problematic internet use in adolescents.

5. Conclusions

The study suggests that parent-reported problematic internet use warrants further investigation among adolescents in Cyprus, while emphasizing that the observed distribution should not be interpreted as a representative population prevalence estimate. In this convenience sample, approximately 74% of adolescents fell within the broader problematic internet use range according to published PCIAT cut-offs used descriptively, not as clinically validated Cypriot thresholds. Separately, 2% of adolescents were classified in the severe PCIAT category. Extended daily use of the internet, poorer family functioning, and externalizing behavioral problems such as conduct issues and hyperactivity/inattention were associated with higher problematic internet use severity, although the modest explained variance indicates that additional unmeasured factors are likely to be important. These findings underscore the need for comprehensive, multidimensional intervention strategies that address both individual digital habits and family dynamics. Future research using longitudinal, representative, socioeconomically diverse, and multi-informant designs is essential to clarify the population prevalence, temporal directionality, and clinically relevant correlates of problematic internet use in Cyprus and beyond.

Author Contributions

Conceptualization, C.P. and G.G.; methodology, C.P. and G.G.; software, C.P.; validation, C.P., G.K. and G.G.; formal analysis, C.P. and G.G.; investigation, C.P.; resources, C.P.; data curation, C.P.; writing—original draft preparation, C.P. and G.G.; writing—review and editing, G.K., K.S. and G.G.; visualization, G.G.; supervision, G.G.; project administration, C.P. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

The study was conducted in accordance with the Declaration of Helsinki and approved by the National Bioethics Committee of Cyprus (Approval No. ΕΕΒΚ ΕΠ 2022.01.143) on 9 June 2022.

Data Availability Statement

The data presented in this study are available on request from the corresponding author due to privacy reasons.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
GF-12General Functioning Scale
PCIATParent–Child Internet Addiction Test
SDQStrengths and Difficulties Questionnaire

References

  1. Haleem, A.; Javaid, M.; Qadri, M.A.; Suman, R. Understanding the role of digital technologies in education: A review. Sustain. Oper. Comput. 2022, 3, 275–285. [Google Scholar] [CrossRef] [Scilit]
  2. Levin, I.; Mamlok, D. Culture and Society in the Digital Age. Information 2021, 12, 68. [Google Scholar] [CrossRef] [Scilit]
  3. Magis-Weinberg, L.; Ballonoff Suleiman, A.; Dahl, R.E. Context, Development, and Digital Media: Implications for Very Young Adolescents in LMICs. Front. Psychol. 2021, 12, 632713. [Google Scholar] [CrossRef] [Scilit]
  4. Marciano, L.; Ostroumova, M.; Schulz, P.J.; Camerini, A.L. Digital Media Use and Adolescents’ Mental Health During the Covid-19 Pandemic: A Systematic Review and Meta-Analysis. Front. Public Health 2021, 9, 793868. [Google Scholar] [CrossRef] [Scilit]
  5. Pagano, M.; Bacaro, V.; Crocetti, E. “Using digital media or sleeping … that is the question”. A meta-analysis on digital media use and unhealthy sleep in adolescence. Comput. Hum. Behav. 2023, 146, 107813. [Google Scholar] [CrossRef] [Scilit]
  6. Moretta, T.; Buodo, G.; Demetrovics, Z.; Potenza, M.N. Tracing 20 years of research on problematic use of the internet and social media: Theoretical models, assessment tools, and an agenda for future work. Compr. Psychiatry 2022, 112, 152286. [Google Scholar] [CrossRef] [Scilit]
  7. Chung, S.; Lee, H.K. Public Health Approach to Problems Related to Excessive and Addictive Use of the Internet and Digital Media. Curr. Addict. Rep. 2023, 10, 69–76. [Google Scholar] [CrossRef] [Scilit]
  8. Fontana, A.; Benzi, I.M.A.; Cipresso, P. Problematic internet use as a moderator between personality dimensions and internalizing and externalizing symptoms in adolescence. Curr. Psychol. 2023, 42, 19419–19428. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  9. Piccerillo, L.; Digennaro, S. Adolescent Social Media Use and Emotional Intelligence: A Systematic Review. Adolesc. Res. Rev. 2025, 10, 201–218. [Google Scholar] [CrossRef] [Scilit]
  10. Rozgonjuk, D.; Blinka, L.; Löchner, N.; Faltýnková, A.; Husarova, D.; Montag, C. Differences between problematic internet and smartphone use and their psychological risk factors in boys and girls: A network analysis. Child Adolesc. Psychiatry Ment. Health 2023, 17, 69. [Google Scholar] [CrossRef] [Scilit]
  11. D’Angelo, J.D. Chapter 16—Problematic and addictive technology use. In Technology and Adolescent Health; Moreno, M.A., Hoopes, A.J., Eds.; Academic Press: Cambridge, MA, USA, 2020; pp. 347–364. [Google Scholar]
  12. Kuss, D.J.; Lopez-Fernandez, O. Internet addiction and problematic Internet use: A systematic review of clinical research. World J. Psychiatry 2016, 6, 143–176. [Google Scholar] [CrossRef] [Scilit]
  13. Restrepo, A.; Scheininger, T.; Clucas, J.; Alexander, L.; Salum, G.A.; Georgiades, K.; Paksarian, D.; Merikangas, K.R.; Milham, M.P. Problematic internet use in children and adolescents: Associations with psychiatric disorders and impairment. BMC Psychiatry 2020, 20, 252. [Google Scholar] [CrossRef] [Scilit]
  14. Young, K.S. Internet addiction: The emergence of a new clinical disorder. CyberPsychol. Behav. 1998, 1, 237–244. [Google Scholar] [CrossRef] [Scilit]
  15. Shek, D.T.L.; Yu, L.; Sun, R.C.F.; Fan, Y. Internet Addiction. In Neuroscience in the 21st Century: From Basic to Clinical; Pfaff, D.W., Volkow, N.D., Rubenstein, J., Eds.; Springer: New York, NY, USA, 2020; pp. 1–42. [Google Scholar]
  16. Hunt, A.; Merola, G.P.; Carpenter, T.; Jaeggi, A.V. Evolutionary perspectives on substance and behavioural addictions: Distinct and shared pathways to understanding, prediction and prevention. Neurosci. Biobehav. Rev. 2024, 159, 105603. [Google Scholar] [CrossRef] [Scilit]
  17. Chang, M.L.; Lee, I.O. Functional connectivity changes in the brain of adolescents with internet addiction: A systematic literature review of imaging studies. PLoS Ment. Health 2024, 1, e0000022. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  18. Sakamoto, S.; Miyawaki, D.; Goto, A.; Hirai, K.; Hama, H.; Kadono, S.; Nishiura, S.; Inoue, K. Associations between Internet Addiction, Psychiatric Comorbidity, and Maternal Depression and Anxiety in Clinically Referred Children and Adolescents. Neuropsychiatr. Dis. Treat. 2022, 18, 2421–2430. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  19. Bickham, D.S. Current Research and Viewpoints on Internet Addiction in Adolescents. Curr. Pediatr. Rep. 2021, 9, 1–10. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  20. Diotaiuti, P.; Girelli, L.; Mancone, S.; Corrado, S.; Valente, G.; Cavicchiolo, E. Impulsivity and Depressive Brooding in Internet Addiction: A Study With a Sample of Italian Adolescents During COVID-19 Lockdown. Front. Psychiatry 2022, 13, 941313. [Google Scholar] [CrossRef] [Scilit]
  21. Aziz, M.; Chemnad, K.; Al-Harahsheh, S.; Abdelmoneium, A.O.; Baghdady, A.; Ali, R. Depression, stress, and anxiety versus internet addiction in early and middle adolescent groups: The mediating roles of family and school environments. BMC Psychol. 2024, 12, 184. [Google Scholar] [CrossRef] [Scilit]
  22. Ding, K.; Li, H. Digital Addiction Intervention for Children and Adolescents: A Scoping Review. Int. J. Environ. Res. Public Health 2023, 20, 4777. [Google Scholar] [CrossRef] [Scilit]
  23. Wang, Y.; Zhou, K.; Wang, Y.; Zhang, J.; Xie, Y.; Wang, X.; Yang, W.; Zhang, X.; Yang, J.; Wang, F. Examining the association of family environment and children emotional/behavioral difficulties in the relationship between parental anxiety and internet addiction in youth. Front. Psychiatry 2024, 15, 1341556. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  24. Miao, S.; Xu, L.; Gao, S.; Bai, C.; Huang, Y.; Peng, B. The association between anxiety and internet addiction among left-behind secondary school students: The moderating effect of social support and family types. BMC Psychiatry 2024, 24, 406. [Google Scholar] [CrossRef] [Scilit]
  25. Huang, X.-c.; Zhang, Y.-n.; Wu, X.-y.; Jiang, Y.; Cai, H.; Deng, Y.-q.; Luo, Y.; Zhao, L.-p.; Liu, Q.-l.; Luo, S.-y.; et al. A cross-sectional study: Family communication, anxiety, and depression in adolescents: The mediating role of family violence and problematic internet use. BMC Public Health 2023, 23, 1747. [Google Scholar]
  26. Procentese, F.; Gatti, F.; Di Napoli, I. Families and Social Media Use: The Role of Parents’ Perceptions about Social Media Impact on Family Systems in the Relationship between Family Collective Efficacy and Open Communication. Int. J. Environ. Res. Public Health 2019, 16, 5006. [Google Scholar] [CrossRef] [Scilit]
  27. Lozano-Blasco, R.; Latorre-Martínez, M.P.; Cortés-Pascual, A. Screen addicts: A meta-analysis of internet addiction in adolescence. Child. Youth Serv. Rev. 2022, 135, 106373. [Google Scholar] [CrossRef] [Scilit]
  28. Wang, E.; Zhang, J.; Dong, Y.; Xiao, J.; Qu, D.; Shan, H.; Chi, X. Vicious circle of family dysfunction and adolescent internet addiction: Do only child and non-only child exhibit differences? Curr. Psychol. 2024, 43, 827–838. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  29. Sheng, X.; Yang, M.; Ge, M.; Zhang, L.; Huang, C.; Cui, S.; Yuan, Q.; Ye, M.; Zhou, R.; Cao, P.; et al. The relationship between Internet addiction and childhood trauma in adolescents: The mediating role of social support. Front. Psychol. 2022, 13, 996086. [Google Scholar] [CrossRef] [Scilit]
  30. Doğrusever, C.; Bilgin, M. From family social support to problematic internet use: A serial mediation model of hostility and depression. BMC Psychol. 2025, 13, 178. [Google Scholar] [CrossRef] [Scilit]
  31. Calaresi, D.; Verrastro, V.; Giordano, F.; Gullo, J.; Saladino, V. Pathways from Family Functioning to Internet Gaming Disorder: The Mediating Role of the Dark Triad. Behav. Sci. 2024, 14, 668. [Google Scholar] [CrossRef] [Scilit]
  32. Siomos, K.; Paradeisioti, A.; Hadjimarcou, M.; Mappouras, D.G.; Kalakouta, O.; Avagianou, P.; Floros, G. The impact of Internet and PC addiction in school performance of Cypriot adolescents. Stud. Health Technol. Inform. 2013, 191, 90–94. [Google Scholar] [PubMed]
  33. Floros, G.; Paradeisioti, A.; Hadjimarcou, M.; Mappouras, D.G.; Kalakouta, O.; Avagianou, P.; Siomos, K. Cyberbullying in Cyprus—Associated parenting style and psychopathology. Stud. Health Technol. Inform. 2013, 191, 85–89. [Google Scholar] [PubMed]
  34. Young, K. Internet Addiction Test for Families (IAT-F); Stoelting: Wood Dale, IL, USA, 2017. [Google Scholar]
  35. Goodman, R. The Strengths and Difficulties Questionnaire: A research note. J. Child. Psychol. Psychiatry 1997, 38, 581–586. [Google Scholar] [CrossRef] [Scilit]
  36. Giannakopoulos, G.; Dimitrakaki, C.; Papadopoulou, K.; Tzavara, C.; Kolaitis, G.; Ravens-Sieberer, U.; Tountas, Y. Reliability and validity of the Strengths and Difficulties Questionnaire in Greek adolescents and their parents. Health 2013, 2013, 1774–1783. [Google Scholar] [CrossRef]
  37. Epstein, N.B.; Bishop, D.S.; Levin, S. The McMaster model of family functioning. J. Marital. Fam. Ther. 1978, 4, 19–31. [Google Scholar] [CrossRef] [Scilit]
  38. Tsamparli, A.; Petmeza, I.; McCarthy, G.; Adamis, D. The Greek version of the McMaster Family Assessment Device. PsyCh J. 2018, 7, 122–132. [Google Scholar] [CrossRef] [Scilit]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Personality, Interest and Career Decisions: Assessing Psychological Readiness for Professional Choice Among High School Students
Previous
The Relationship Between Resilience, Self-Esteem, and Academic Performance: An Investigation in Primary School Students
Next