INTRODUCTION
Smartphone use has become nearly universal among adolescents worldwide, with Korean youth reporting some of the highest rates of device ownership and daily screen time [1,2]. While smartphones offer clear benefits for communication, information access, and education, a growing body of research has identified problematic smartphone use (PSU) as an emerging behavioral concern [3,4]. PSU is typically characterized by excessive, poorly controlled use that interferes with daily functioning, and it is accompanied by distress, craving, or impairment in social, academic, or emotional domains [5-7]. Studies across different regions have linked PSU to a range of adverse outcomes, including sleep disturbance, depressive symptoms, anxiety, and self-harm behaviors, among adolescents [2,8].
In South Korea, national survey data indicate that a substantial proportion of adolescents are at risk for smartphone overdependence, with recent estimates suggesting that approximately one-fourth to one-third of youth fall into potentialrisk or highrisk categories [2,9,10]. These rates are comparable to or slightly higher than pooled estimates from international meta-analyses, which place adolescent PSU prevalence at approximately 20%-30% [11,12]. Korean adolescents are exposed to a confluence of contextual pressures, including intense academic competition, high levels of digitalization, and strong social expectations regarding online connectivity, that may increase vulnerability to problematic digital behaviors [13]. Within this context, PSU has been associated with internalizing symptoms, suicidal ideation, and academic impairment in representative Korean samples [9].
Adolescents from multicultural families in Korea constitute a growing population that faces distinct psychosocial challenges. Previous research has consistently shown that compared with their nonmulticultural peers, multicultural youth are more likely to report depressive symptoms, anxiety, and suicidality, as well as social difficulties such as bullying, discrimination, and school maladjustment [14-16]. These vulnerabilities are often embedded in broader structural disadvantages, including lower socioeconomic status, language barriers, and limited access to culturally responsive support services [16,17]. As a result, multicultural adolescents may be particularly inclined to use smartphones and online environments as spaces for escape, social connection, or identity exploration, which could both buffer and exacerbate their mental health risks [3,18].
Despite the convergence of digital and psychosocial vulnerabilities, relatively few empirical studies have examined PSU among multicultural adolescents. Existing work on Korean youth has focused primarily on PSU in the general adolescent population, without specifically examining multicultural subgroups or exploring how cultural and socioeconomic contexts may shape the association between PSU and mental health [19,20]. Moreover, although prior research has documented bivariate links between PSU and individual outcomes such as depression or suicidality, less is known about how PSU relates to multiple mental health indicators simultaneously or whether these associations differ across key subgroups (e.g., by sex, household income, or levels of stress and loneliness) [16,17,21,22]. Addressing these gaps is important for informing targeted, culturally sensitive prevention and intervention strategies.
The present study sought to extend the literature by examining PSU among Korean adolescents from multicultural families using nationally representative data from the 2020 Korea Youth Risk Behavior Web-based Survey. First, we estimated the prevalence of PSU in this population and situated it within the context of national estimates for Korean adolescents overall. Second, we examined the associations between PSU and multiple mental health indicators—perceived stress, loneliness, anxiety symptoms, depressive mood, and suicidal behaviors—as well as academic achievement. Third, using structural equation modeling with moderation, we investigated whether the strength of the associations between PSU and mental health outcomes varied according to sociodemographic and psychosocial characteristics, including sex, household income, academic achievement, perceived stress, and loneliness. By integrating these objectives, this study provides a more comprehensive picture of how PSU co-occurs with mental health risks among multicultural adolescents and identifies subgroups that may benefit most from early identification and tailored intervention.
METHODS
Study population and data collection
The data for this study came from the 2020 Korean Youth Risk Behavior Survey (KYRBS), a nationwide school-based survey conducted annually by the Korea Disease Control and Prevention Agency [23,24]. The KYRBS uses an anonymous, self-administered online questionnaire and employs a complex sampling design that incorporates stratification, clustering, and multistage probability sampling to obtain a nationally representative sample of Korean middle and high school students [9]. The 2020 survey was conducted between August and November and included 57,925 adolescents from 800 schools, yielding a participation rate of 94.9%. In the KYRBS dataset, multicultural status is defined based on parental country of birth, where adolescents are classified as belonging to multicultural families if at least one parent was born outside Korea. Of the participants in the 2020 survey, 1,065 respondents were identified as belonging to multicultural families and were included in the present analyses (mean age=14.39 years, standard deviation [SD]=1.72; 46.1% boys).
Measurement
PSU
PSU was assessed via the Korean version of the Smartphone Overdependence Scale, which has been validated for use in Korean adolescents [25]. The scale consists of 10 items rated on a 4point Likert scale (1=“not at all,” 2=“disagree,” 3=“agree,” and 4=“always”), yielding total scores ranging from 10 to 40; higher scores indicate greater smartphone overdependence [26,27]. In line with the developer’s guidelines and national surveillance practices, the scores were categorized into three levels: general use (≤22), potential risk (23-30), and high risk (≥31). For the present analysis, adolescents in the potential-risk and high-risk categories were combined and classified into the PSU group, and those in the general-use category were classified into the non-PSU group, which is consistent with previous KYRBS-based studies that used this scale. The reliability of the scale used in this study was 0.89.
Mental health variables
Perceived stress
Perceived stress was measured with the question “How do you generally perceive your stress?,” and there were five response options (“very severe,” “severe,” “moderate,” “little,” and “never”). Following established practices in Korean adolescent surveillance and prior studies that used KYRBS data, responses of “very severe” or “severe” were coded as high stress, and responses of “moderate,” “little,” or “never” were coded as not high stress [28]. This dichotomization facilitates comparison with previous national reports and reflects conventional cut-offs in this data source [24].
Loneliness
Loneliness was assessed with the item “How often did you feel lonely during the past 12 months?.” Responses were provided on a 5-point scale ranging from “never” to “always.” In line with prior KYRBS analyses and other adolescent mental health research that use similar single-item measures, we dichotomized the responses into frequent versus infrequent loneliness: “always” and “often” were coded as frequent loneliness, whereas “sometimes,” “rarely,” and “never” were coded as infrequent loneliness. This approach is commonly used to distinguish adolescents who experience substantial, recurrent loneliness from those with lower levels of lonely feelings.
Depressive mood
Depressive mood was measured with the question “During the past 12 months, have you ever felt so sad or hopeless that you stopped your usual activities for at least two weeks?” which had yes/no response options. This item is widely used in adolescent surveillance to capture clinically relevant depressive mood [24].
Anxiety
Anxiety symptoms were assessed via the 7-item Generalized Anxiety Disorder (GAD-7) scale [29]. The GAD-7 is a self-report screening instrument that evaluates the severity of anxiety symptoms over the preceding 2 weeks and is not a diagnostic tool. Each item is rated from 0 (“not at all”) to 3 (“nearly every day”), yielding total scores between 0 and 21; higher scores indicate more severe anxiety symptoms. Based on conventional cut-offs, total scores were categorized as no/minimal (0-4), mild (5-9), moderate (10-14), and severe (15-21) anxiety. In the present study, consistent with previous adolescent research using the GAD-7, participants were recategorized into two groups for analysis: a general group (no/minimal anxiety) and an anxiety group (mild, moderate, or severe anxiety). This dichotomization was applied to ensure adequate cell sizes, enhance statistical power, and align with prior epidemiological studies that treated any anxiety above the minimal range as clinically meaningful in adolescent populations. The reliability of the GAD-7 scale used in this study was 0.86.
Suicidal behaviors
Suicidal behaviors were assessed with three yes/no questions asking whether, during the past 12 months, the adolescent had seriously considered suicide (suicidal ideation), made a specific plan for suicide (suicidal planning), or attempted suicide (suicide attempt). These items have shown acceptable reliability in previous research that used national adolescent survey data [9].
Covariate variables
The covariates included sex (boys or girls), area of residence (metropolitan, urban, or rural), the school type (middle or high school), household income (high, middle, or low), and self-rated academic achievement (high, middle, or low). We also adjusted for health risk behaviors, specifically lifetime cigarette use (yes/no) and alcohol use (yes/no) [24].
Statistical analysis
Statistical analyses were conducted using survey procedures that accounted for the complex sampling design of the KYRBS, including stratification, clustering, and sampling weights [30]. Descriptive statistics were used to summarize the sociodemographic and psychosocial characteristics of the sample. The group differences between adolescents with and without PSU were examined via chi-square tests for categorical variables and independent-samples t-tests for continuous variables. To identify factors associated with PSU, we fitted multivariable logistic regression models adjusting for sex, area of residence, school type, household income, academic achievement, cigarette use, and alcohol use. The results are presented as adjusted odds ratios with 95% confidence intervals.
We then applied structural equation modeling (SEM) to examine the associations between PSU and multiple mental health indicators (perceived stress, loneliness, anxiety, depressive mood, suicidal planning, and academic achievement) within a single analytic framework while controlling for the same covariates. SEM was used to simultaneously evaluate the relationships between PSU and multiple mental health indicators while accounting for potential moderating effects. Moderation by key sociodemographic and psychosocial factors (e.g., sex, household income, perceived stress, loneliness, and academic achievement) was evaluated by including interaction terms and, where appropriate, by comparing subgroup models. Overall model fit was evaluated via standard indices, including the comparative fit index, Tucker-Lewis index, root mean square error of approximation, and standardized root mean square residual. Statistical significance was defined as a two-sided p-value<0.05, and statistical analyses were conducted using IBM SPSS Statistics (version 25.0; IBM Corp.).
Ethics statement
This study analyzed de-identified data from the 2020 KYRBS, a nationally representative survey conducted by the Korea Disease Control and Prevention Agency (approval No. 117058). The Institutional Review Board of the College of Medicine, The Catholic University of Korea, waived the requirement for individual informed consent and approved the study (IRB No. MC22ZISI0048).
RESULTS
Sample characteristics
A total of 1,065 multicultural adolescents were included in the analysis, and their characteristics are presented in Table 1. The mean age was 14.39 years (SD=1.72), and just over half were girls. Most participants reported a middle-level household income and middle-level academic achievement. A large proportion reported high levels of perceived stress and loneliness, and notable subsets met the screening threshold for anxiety symptoms or reported suicidal ideation. PSU was identified in approximately one quarter of the sample.
Group comparisons by PSU status
Sociodemographic and psychosocial characteristics by PSU status are summarized in Table 2 and Figure 1. Compared with their non-PSU peers, adolescents with PSU were more likely to report lower academic achievement, higher perceived stress, more frequent loneliness, elevated anxiety symptoms, and greater involvement in suicidal ideation and planning. In contrast, there were no clear differences between the PSU and non-PSU groups in terms of residential area, school type, household income category, or cigarette and alcohol use. Overall, the pattern of the bivariate results indicates that PSU tends to cluster with multiple indicators of psychosocial vulnerability among multicultural adolescents, whereas basic demographic characteristics are more evenly distributed across groups.
Multivariable logistic regression
Table 3 and Figure 2 present the multivariable logistic regression models used to examine the factors associated with PSU. After adjusting for sex, residential area, school type, household income, academic achievement, cigarette use, and alcohol use, PSU remained significantly related to several mental health indicators. Adolescents reporting high levels of perceived stress, frequent loneliness, or anxiety symptoms had higher odds of belonging to the PSU group. Lower academic achievement and lower household income were also associated with increased odds of PSU. In contrast, sex, residential area, school type, cigarette and alcohol use, depressive mood, suicidal ideation, and suicide attempts were not significantly associated with PSU once the covariates were taken into account. These findings suggest that PSU is more closely linked to psychosocial distress and academic difficulties than to basic demographic factors or general health-risk behaviors.
SEM and moderation
The structural equation models and moderation analyses are summarized in Table 4 and Figure 3. Higher PSU scores were positively associated with perceived stress, loneliness, anxiety symptoms, and suicidal planning and negatively associated with academic achievement after the covariates were accounted for. No statistically significant direct association between PSU and depressive mood was observed. Moderation analyses indicated that some of these associations were stronger in specific subgroups; for example, the links between PSU and loneliness were more pronounced among girls, the associations between PSU and anxiety symptoms were stronger among adolescents from lower-income households, and the association between PSU and suicidal planning was heightened among those reporting higher levels of stress or loneliness. The overall model showed a good fit to the data. Taken together, these results suggest that PSU is embedded in a broader network of psychosocial difficulties and that certain groups of multicultural adolescents may experience particularly elevated risk.
DISCUSSION
The present study identified a 26.2% prevalence of PSU among Korean adolescents from multicultural families, which is a level broadly that is consistent with national estimates for Korean adolescents overall and with international meta-analytic findings in similar age groups [19,21]. This consistency suggests that PSU is not a rare phenomenon; rather, it is a common behavioral pattern within this population, with potential implications for both mental health and academic functioning [13,21]. Given the rapid expansion of digital media use among youth in Korea, these findings underscore the need to systematically monitor PSU as part of routine adolescent health surveillance [22,31].
The SEM findings suggest that PSU is linked to multiple psychological difficulties simultaneously, which highlights the interconnected nature of digital behavior and mental health among adolescents. Consistent with prior work in general adolescent populations, PSU among multicultural adolescents was associated with higher levels of perceived stress, more frequent loneliness, elevated anxiety symptoms, and increased suicidal planning [18,32]. These associations remained significant after adjustment for sociodemographic and behavioral covariates, suggesting that PSU co-occurs with a broader profile of psychosocial vulnerability rather than representing an isolated behavior [33]. At the same time, the cross-sectional design does not allow us to determine the temporal ordering of these associations; it is plausible that psychological distress contributes to an increased reliance on smartphones, that PSU exacerbates distress, or that both processes operate simultaneously [5]. Future longitudinal and experimental research will be crucial for disentangling these pathways and clarifying potential underlying mechanisms.
The absence of a direct adjusted association between PSU and depressive mood in the structural equation models contrasts with the findings of some previous studies that have reported robust links between PSU and depression among adolescents [13,14,16]. One explanation may be that among multicultural adolescents, depressive mood is more strongly shaped by contextual factors such as family functioning, discrimination, and chronic socioeconomic adversity, which were not fully captured in the present dataset [11,34]. It is also possible that the association between PSU and depressive symptoms is mediated by intermediate constructs, including perceived stress, loneliness, or sleep disturbance, which warrants examination in future mediation analyses [12,35,36]. These findings highlight the importance of considering multiple indicators of emotional distress rather than focusing solely on depressive mood when characterizing the mental health correlates of PSU.
Our moderation analyses further suggest that the correlates of PSU are not uniform across all multicultural adolescents. The associations between PSU and loneliness were stronger among girls, while the link between PSU and anxiety symptoms was more pronounced among adolescents from lower-income households, and the association between PSU and suicidal planning was heightened among those reporting higher levels of stress or loneliness. These patterns indicate that PSU is embedded in a broader network of risk factors and that certain subgroups, such as girls, youth from socioeconomically disadvantaged households, and adolescents experiencing high psychosocial strain, may be particularly vulnerable to the adverse correlates of problematic digital engagement [37-39]. Although causal inferences cannot be drawn, these subgroup differences point to the need for tailored assessments and interventions that address both digital behaviors and the underlying psychosocial context.
Given the increasing proportion of multicultural families in South Korea, understanding the digital behaviors and associated mental health risks among this population is an emerging priority for adolescent mental health research and policy. Importantly, our findings do not imply that the observed patterns are unique to multicultural adolescents or that multicultural status itself is the causal driver of PSU or poor mental health. Associations between PSU, internalizing symptoms, loneliness, and suicidality have been documented in many adolescent samples, including those without multicultural backgrounds [9,36]. Rather, our results suggest that multicultural adolescents may experience a convergence of general adolescent risks and additional challenges related to cultural adaptation, minority status, and socioeconomic strain. In this sense, multicultural status can be viewed as a context in which widely observed PSU-mental health associations may be amplified rather than as a singular explanatory factor.
From a clinical and public health perspective, the combination of a high PSU prevalence and its associations with stress, loneliness, anxiety, suicidal planning, and lower academic achievement supports the need for early, context-sensitive intervention [40,41]. School-based screening for PSU and emotional distress, combined with psychoeducation on healthy digital use and coping strategies, may help identify adolescents who are struggling and facilitate timely referral [42]. For multicultural adolescents and their families, interventions that address language barriers, experiences of discrimination, and access to supportive relationships may be especially important, as these factors are likely to interact with digital behaviors to shape mental health outcomes [43-45]. Collaboration among schools, community services, and health care providers is essential to ensure that preventive efforts are both culturally responsive and sustainable [37,46].
Several limitations should be considered when interpreting these findings. First, because the data are cross-sectional, the directionality of the observed associations cannot be determined; psychological distress may contribute to increased smartphone use, PSU may exacerbate psychological difficulties, or both processes may occur simultaneously. Second, all measures were based on self-reported responses, which may introduce reporting bias or social desirability effects. Third, the dichotomization of perceived stress and loneliness, although consistent with national surveillance practices and prior KYRBS-based studies, may have reduced the variability in these constructs and should be interpreted as a methodological limitation. Fourth, this study focused exclusively on multicultural adolescents and did not include a statistically comparable control group of nonmulticultural adolescents, which precludes direct group comparisons and limits our ability to determine whether the observed associations differ in magnitude from those seen in the general adolescent population. Future research using explicitly comparative designs that incorporate both multicultural and nonmulticultural youth is needed to clarify these differences. Fifth, the measures did not capture several potentially important contextual factors, such as family dynamics, acculturative stress, experiences of discrimination, or sleep patterns, which could help explain the observed associations. Finally, although the sample is nationally representative of Korean adolescents, the findings may not be generalizable to multicultural youth in other countries with different educational systems, immigration histories, and social policies. Despite these limitations, the consistent pattern of associations across multiple indicators supports the robustness of the observed links between PSU and psychosocial risks among multicultural adolescents.
Conclusion
This study provides robust evidence that PSU is independently linked to elevated risks of anxiety symptoms, perceived stress, loneliness, and suicidal planning among multicultural adolescents in South Korea. These results underscore the presence of distinct psychosocial vulnerabilities in this population, and they emphasize the need for culturally tailored strategies to address the mental health consequences of digital behaviors. Future longitudinal research is needed to explain the underlying mechanisms and to inform the development of evidence-based public health policies that promote the wellbeing of multicultural youth in increasingly digital societies.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Print
