Skip to main content
NCBI home page
BMC Public Health logoLink to BMC Public Health
. 2026 Apr 6;26:1574. doi: 10.1186/s12889-026-27258-8

Mental health problems as mediators between self-esteem and internet addiction among adolescents

Dat Tan Nguyen 1, Thuy Doan Hua 2,3, Thao Thanh Thi Nguyen 1,4,✉, Nhan Hong Pham 5,6, Huyen Ngoc Nguyen 1, Cyrus Su Hui Ho 7,8, Vinh-Long Tran-Chi 3,9, Kristine Lin 10, Chung-Ying Lin 4,11,12,13,14,✉
PMCID: PMC13192003  PMID: 41937156

Abstract

Background

Mental health problems (i.e., depression, anxiety, stress) experienced by adolescents may be associated with internet addiction. Self-esteem may be negatively related to mental health problems, especially in teenagers. Self-esteem may be indirectly associated with internet addiction through its relationships with adolescents’ mental health problems. Therefore, this study aimed to explore mental health problems that mediated the association between self-esteem and internet addiction.

Methods

A cross-sectional study was carried out between February and March 2025 among Vietnamese high school students. Participants filled out questionnaires on demographic characteristics, the Depression Anxiety Stress Scale-21, the Rosenberg Self-Esteem Scale, and the Internet Addiction Test-Short Version. Group differences were examined using nonparametric tests (Kruskal-Wallis and Mann-Whitney U tests), while partial least squares structural equation modeling (PLS-SEM) with 5,000 bootstrap resamples was the primary analytical approach for testing mediation hypotheses and estimating path coefficients.

Results

In the total sample of 789 Vietnamese high school students (mean age = 16.96 years, standard deviation = 0.80), 54.6% were female. Depression (β = 0.180, p = .002), anxiety (β = 0.214, p < .001), and stress (β = 0.306, p < .001) were positively associated with internet addiction. Self-esteem was negatively associated with depression (β = -0.568, p < .001), anxiety (β = -0.537, p < .001), stress (β = -0.400, p < .001), and internet addiction (β = -0.288, p < .001). Additionally, depression (β = -0.102, p = .003), anxiety (β = -0.115, p < .001), and stress (β = -0.122, p < .001) showed significant indirect associations between self-esteem and internet addiction.

Conclusion

The study provides insight into the associations between self-esteem and internet addiction, considering the mediating roles of mental health problems. While causal relationships cannot be drawn, the findings highlight the potential importance of these psychological factors in shaping internet use behaviors. Internet addiction reduction programs may thus consider incorporating mental health and self-esteem components into youth-focused awareness initiatives.

Keywords: Self-esteem, Depression, Anxiety, Stress, Internet addiction, Adolescents, High school students

Introduction

Adolescence, defined as the period from ages 10 to 19, is a critical transitional phase marked by an increase in mental health problems in both complexity and severity [59, 84]. Worldwide, around 7 out of 50 adolescents experience mental health problems [85], often leading to poor academic performance, dropout, family conflict, and risky behaviors [45, 55, 58, 59]. Apart from mental health problems (i.e., depression, anxiety, and stress in the present study), internet addiction has emerged as a prominent psychological concern among youths [1, 29, 56, 64, 67]. Internet addiction is defined as patterns of internet behavior that cause psychological, academic, social, or occupational difficulties in individuals’ lives [70]. However, prevalence data on mental health problems among adolescents in less economically developed countries remain limited and often lack generalizability [15, 39, 45, 46].

While previous studies have examined the relationships among self-esteem, depression, anxiety, stress, and internet addiction [39, 54], few studies have explicitly investigated the mediating role of mental health problems (i.e., depression, anxiety, and stress) simultaneously within a single model among adolescents in our cultural context, particularly in Vietnam. Vietnam represents a unique sociocultural and economic setting, characterized by rapid socioeconomic transitions, strong collectivistic values, intergenerational family structures, and persistent stigma surrounding mental health [60]. Moreover, Vietnam, as a low- and middle-income country (LMIC), is currently facing growing mental health challenges, including increasing rates of depression, anxiety, and stress-related problems, while mental health services and resources remain limited [54, 59]. Therefore, understanding how these variables operate within the Vietnamese population is crucial for informing culturally appropriate prevention and intervention strategies.

The relationships between mental health problems and internet addiction

Mental health problems have shown strong associations with internet addiction [68]. Several theoretical frameworks provide insights into how mental health problems are related to internet addiction, including the Interaction of Person-Affect-Cognition-Execution (I-PACE), social support, and coping models [7, 17, 77]. Among them, the I-PACE model posits that the interactions between personal characteristics and emotional, cognitive, and executive processes shape the emergence and persistence of problematic internet use [7]. Furthermore, according to social support and coping theories, people with mental health problems could turn to online activities as a coping mechanism to manage stressors and issues in their lives [77]. As a result of their inability to manage the use of online activities as helpful coping methods, people can suffer internet-related addiction [7].

The relationships between self-esteem and mental health problems

Self-esteem is a crucial psychological construct closely linked to adolescent mental health [51, 54, 66]. Previous studies indicate that adolescents with higher self-esteem generally report lower feelings of loneliness, lower levels of mental health problems, better emotional well-being, and greater overall life satisfaction [49, 50, 57, 62, 66]. In particular, longitudinal studies showed that low self-esteem predicted mental health problems over time [41, 42].

In the vulnerability model of self-esteem, individuals with low self-esteem are inclined to perceive stressful events as personal failures, amplifying their mental health problems [62]. According to the sociometer theory, self-esteem is a psychological factor that fosters a sense of social belonging, enhances social connectedness, and strengthens perceived social support, which in turn may reduce adolescents’ vulnerability to internalizing symptoms [74]. Anxiety mediates the link between self-esteem and internet addiction, as individuals with low self-esteem and high anxiety may use the internet as a compensatory strategy to reduce stress and seek psychological safety [35]. Therefore, these findings underscore the importance of fostering self-esteem as a potential intervention target in mental health prevention and treatment [83].

The relationship between self-esteem and internet addiction

Self-esteem has been consistently identified as an important psychological factor associated with internet addiction among adolescents. Numerous studies indicate that lower self-esteem is associated with higher levels of internet addiction [3, 6, 10, 27, 54]. Stress and coping strategies, in conjunction with self-esteem, are linked to adolescents’ susceptibility to internet addiction, highlighting a complex interplay of psychological factors [2]. A recent study during the COVID-19 pandemic confirmed that adolescents with lower self-esteem experienced increased internet use and addiction, exacerbated by social isolation [78]. In addition, research focusing on social media use among adolescents has shown that individuals with low self-esteem tend to spend more time on these platforms [72].

Other potential factors associated with self-esteem, mental health problems, and internet addiction

Gaining students’ academic and psychological development requires understanding how their background variables (including residential areas, feelings toward school environment, happiness, and family history of psychotic diseases or alcohol addiction) associate with their self-esteem, mental health problems, and internet addiction. For example, urban students often face greater exposure to digital technologies, which may increase the risk of internet addiction [9]. Positive feelings toward the school environment and happiness are associated with lower rates of mental health problems, whereas adverse environments may heighten emotional distress [4, 8, 48, 52]. The family history of psychotic diseases is associated with higher risks of internet addiction and mental health problems [18, 39, 43]. Moreover, self-compassion, self-care, perfectionistic self-presentation, and entitlement have been related to mental health problems [32, 33]. These factors highlight the complex interplay of environmental, emotional, and familial influences on adolescent well-being, underscoring the need for targeted interventions and informed educational policy.

Study purpose and hypotheses

The potential mediating role of mental health problems in the relationship between self-esteem and internet addiction needs empirical evidence to support, especially among Vietnamese adolescents. To address this gap, the present study’s purpose is to investigate the relationships among self-esteem, mental health problems, and internet addiction, with a specific focus on whether mental health problems mediate the relationships between self-esteem and internet addiction among adolescents.

The present study thus proposes the following primary hypotheses:

  • H1: Self-esteem is negatively associated with depression (H1a), anxiety (H1b), stress (H1c), and internet addiction (H1d).

  • H2: Depression (H2a), anxiety (H2b), and stress (H2c) are positively associated with internet addiction.

  • H3: Depression (H3a), anxiety (H3b), and stress (H3c) significantly mediate the relationship between self-esteem and internet addiction in a negative direction.

In addition to these hypotheses, the present study explored how sociodemographic factors (i.e., residential areas, feelings toward school environment, happiness, and the family history of psychotic diseases or alcohol addiction) of participants associated with self-esteem, mental health problems, and internet addiction.

Methods

Study design and participants

A descriptive cross-sectional design was employed to examine the associations and potential mediation among psychological variables in a large sample of adolescents. Data collection period (February to March 2025) avoided the period of final examination because such a period may temporarily increase mental health problems among Vietnamese high school students. Convenience sampling was used because probability-based sampling was not feasible due to administrative and consent constraints in the school setting. This approach was suitable for this cross-sectional study, which focused on examining associations and mediation rather than population-level estimates.

Eligible participants were Vietnamese students (grades 10 to 12) from two selected schools in Can Tho city, Vietnam, who could comprehend and complete the questionnaires. Participants under 18 years of age provided written assent, and written informed consent was obtained from their parents or legal guardians prior to data collection. Parental consent forms were distributed to participants to be completed by their parents or legal guardians at home, and only those who returned signed forms were permitted to participate. Participants aged 18 years and older provided written informed consent. Exclusion criteria included students who had discontinued their studies, lacked parental consent, or had documented disabilities or severe psychological disorders. Furthermore, questionnaires with missing or incomplete responses were excluded from the final dataset.

Data collection was conducted with administrative support from Can Tho University of Medicine and Pharmacy, Vietnam. Before recruitment, meetings were arranged with school principals in selected high schools in Can Tho City to outline the research objectives and obtain formal consent to approach students. Students in grades 10 to 12 were invited to participate through printed materials distributed during recess and extracurricular periods. Explanatory statements were issued to students and guardians ten days prior to data collection, outlining the study’s aims, procedures, ethical considerations, and support services. During the scheduled data-collection period, the research team visited the selected classrooms during prearranged sessions. Students were reminded of the study’s objectives, their rights to participate, potential risks, and the consent procedure. After collecting assent and consent forms, students spent approximately 20 to 25 min completing the questionnaire via Google Forms.

Sample sizes

According to Hair [24], a sample of 100 to 200 participants is typically sufficient for structural equation models of low to moderate complexity, such as those with around five latent variables and seven structural paths. In the current study, the sample of 789 participants substantially exceeds this recommended range, indicating that it is adequate for obtaining reliable model estimates. The suitability of the sample was assessed using rule-of-thumb ratios of cases to estimated parameters [36]. The model incorporated approximately seven estimated structural parameters (direct effects), while indirect effects were not treated as separate parameters because they are derived from the direct paths. Consequently, the total number of estimated parameters in the model (N = 43) accounts only for direct paths; indirect effects are examined post-estimation, typically via bootstrapping, without increasing the number of free parameters. Using the 5:1 criterion, the minimum sample size required would be 5 × 43 = 215 participants, whereas the more conservative 10:1 criterion would suggest at least 430 participants [36]. The sample size of this study (n = 789) exceeds the minimum requirement recommended for PLS-SEM analysis based on the number of structural paths in the model (ratio = 18.4).

Measurements

Background variables

Our study consisted of several background variables, including age, sex (male and female), grade (10th to 12th ), residential areas (urban and rural areas), feelings toward school environment (very positive and positive, normal, very negative and negative feelings), happiness (very happy and happy, normal, less happy and not happy), and the medical history of family members experiencing psychotic diseases or alcohol addiction (yes and no).

Rosenberg self-esteem scale

Self-esteem was assessed using the 10-item Rosenberg Self-Esteem Scale (RSES), with responses on a 4-point Likert scale (0 = strongly agree to 3 = strongly disagree). The Rosenberg Self-Esteem Scale (RSES) consists of 10 items rated on a 4-point Likert scale ranging from 0 indicating “Strongly disagree” to 3 indicating “Strongly agree”. Five negatively worded items (2, 5, 6, 8, and 9) were reverse-coded. Total scores range from 0 to 30, with higher scores indicating higher self-esteem [65]. The RSES has demonstrated strong reliability, with test–retest correlation coefficients ranging from 0.82 to 0.88. A study conducted in Vietnam by Nguyen et al. [54] found the scale to have acceptable internal consistency (Cronbach’s alpha = 0.77). Items 2, 5, 6, 8, and 9 are reversed before summing up the RSES total score [65]. In our research, Cronbach’s alpha was 0.83, indicating good reliability.

Depression anxiety stress scale-21

The Depression, Anxiety, and Stress Scale-21 (DASS-21), a shortened version of the DASS-42, is commonly used as a screening scale to assess symptoms of depression, anxiety, and stress in community settings [44]. This measure consists of three subscales assessing depression, anxiety, and stress separately. Each DASS-21 subscale consists of 7 items assessing the extent to which the statement has been applied in the past week. There are 21 items in this scale. Each item is rated on a 4-point Likert scale ranging from 0 to 3, with 0 indicating “Did not apply at all – Never” to 3 indicating “Applied very much or most of the time – Almost always”. Responses to all relevant items are summed, and the total is doubled to evaluate levels of depression, anxiety, and stress [79]. The original English-language version of the DASS-21 was psychometrically validated for use with Vietnamese adolescents [40]. The validation process supported the scale’s convergent validity and factor structure, and the scale exhibited high internal consistency, with Cronbach’s alpha of 0.88 for the total score and Cronbach’s alphas for the subscales ranging from 0.70 to 0.77 [40]. Our study reported high internal consistency, with a Cronbach’s alpha of 0.91 for the total score; and 0.84 for depression, 0.79 for anxiety, and 0.78 for stress subscales.

Internet addiction test-short version

The Internet Addiction Test (IAT), originally developed by Young [89], consists of 20 items modeled on the DSM-IV criteria for pathological gambling. It measures various dimensions of problematic internet use, including compulsive behavior, disruption to daily functioning, and withdrawal symptoms. The Internet Addiction Test-Short Version (IAT-SV) consists of 12 items, each rated on a 5-point Likert scale from 1 (rarely) to 5 (always). Consequently, total scores range from 12 to 60. In Vietnam, the IAT-SV has been culturally adapted and psychometrically validated, with a reliability coefficient (Cronbach’s alpha) of 0.87 and a factor structure delineating time management loss and social consequences [80, 81]. The Vietnamese version of the IAT-SV demonstrated strong internal consistency, with a Cronbach’s alpha of 0.87 [80]. In the present study, its internal consistency was good (Cronbach’s alpha = 0.88).

Data analysis

Data entry was performed using Excel, and statistical analyses were conducted using SPSS version 24. Descriptive statistics summarized categorical variables by frequency and percentage. The data across demographic variables were not normally distributed; therefore, Kruskal-Wallis (for three-group comparison) and Mann-Whitney U tests (for two group comparison) were used to evaluate group differences in internet addiction, mental health problems, and self-esteem across related factors (sex, residential areas, feelings toward school environment, happiness, and the medical history of family members experiencing alcohol addiction, psychotic diseases). SmartPLS 4 software was used to test measurement models and structural relationships via partial least squares-structured equation modeling (PLS-SEM). In order to satisfy the principle of parsimony in statistics, the present study chose to use PLS-SEM over covariance-based SEM. Specifically, PLS-SEM is a composite-based modeling approach with fewer estimations than covariance-based SEM (a factor-based modeling that requires the estimation of residuals) [12]. The PLS-SEM constructed in the present study did not control any demographic and psychosocial variables (i.e., sex, age, happiness, and school-related feelings) to satisfy the principle of parsimony as well. The demographic and psychosocial variables were not controlled as we intend to provide clear and simple understandings of the primary studied concepts in the present study. Additionally, this approach emphasizes heterotrait–monotrait (HTMT), explained variance (R²), cross-validated redundancy (Q2), effect size (f2), variance inflation factor (VIF), and standardized root mean square residual (SRMR), which is consistent with the applied and intervention-oriented focus of the study. Reliability and validity were assessed through indicator loadings, construct reliability (CR), and average variance extracted (AVE). Given its suitability for sample sizes and complex models, PLS-SEM was applied with 5,000 bootstrap samples to test mediation hypotheses and estimate path coefficients. Background variables were analyzed separately to describe group differences and were not modeled as covariates in the PLS-SEM, which focused on testing the hypothesized mediation pathways.

Results

Demographic characteristics

The sample (mean [SD] age = 16.96 [0.80] years; age range from 16 to 20) consisted of 431 females (54.6%) and 358 males (45.4%). Participants were high school students aged 16–20 years, the majority of whom fell within the adolescent age range. Participants aged 17 represented the largest proportion of the sample (36.9%), followed by those aged 16 (33.7%) and those aged 18–20 (29.4%). Regarding grade level, 34.6% of students (n = 273) were in tenth grade, 36.6% (n = 289) in eleventh grade, and 28.8% (n = 227) in twelfth grade.

Nonparametric test of group differences

Internet addiction differed significantly across participants in the less and not happy group (33.16 ± 11.36) and those in the very happy and happy group (27.92 ± 9.23; p < .01). Self-esteem differed significantly across participants in the less and not happy group (13.30 ± 5.34) and those in the very happy and happy group (17.82 ± 4.66; p < .01). Internet addiction differed significantly across participants in the very negative and negative feelings toward school environment group (33.59 ± 12.83) and those in the very positive and positive feelings toward school environment group (27.66 ± 8.64; p < .001). Self-esteem differed significantly across participants in the very negative and negative feelings toward school environment group (14.49 ± 5.47) and those in the very positive and positive feelings toward school environment group (17.28 ± 4.49; p < .001), outlined in Table 1.

Table 1.

Differences in Psychological Factors Based on Happiness Levels Group and Feeling toward School Environment

Variables Mean (SD) Chi-square p
Happiness levels

Less and

not happy

 Normal Very happy and happy
Depression 10.60 (4.81) 7.43 (3.69) 5.75 (3.83) 78.78 < .001
Anxiety 10.34 (4.37) 7.79 (3.67) 6.48 (3.77) 52.41
Stress 12.06 (4.43) 9.15 (3.44) 8.01 (3.73) 51.56
Internet addiction 33.16 (11.36) 29.40 (9.07) 27.92 (9.23) 12.99 < .01
Self-esteem 13.30 (5.34) 15.54 (4.25) 17.82 (4.66) 67.16 < .001
Feelings toward the school environment
Very negative and negative Normal Very positive and positive
Depression 8.90 (4.98) 7.16 (4.05) 6.00 (3.75) 31.72 < .001
Anxiety 8.68 (4.73) 7.46 (3.62) 6.82 (3.93) 13.54
Stress 10.08 (4.68) 9.00 (3.62) 8.25 (3.76) 14.35
Internet addiction 33.59 (12.83) 29.39 (9.32) 27.66 (8.64) 16.63 < .001
Self-esteem 14.49 (5.47) 16.33 (4.83) 17.28 (4.49) 19.76 < .001
Open in a new tab

Female students reported higher levels of depression, anxiety, stress, and internet addiction, whereas male students reported higher levels of self-esteem. Depression, anxiety, stress, self-esteem, and internet addiction were significantly higher among participants with a family history of psychotic diseases, compared to those without a family history of psychotic diseases. Depression, anxiety, and stress were significantly higher among participants with a family history of alcohol addiction, compared to those without a family history of alcohol addiction. Moreover, self-esteem was significantly lower among participants living in rural areas than among those living in urban areas (Table 2).

Table 2.

Differences in psychological factors based on sex, family history, and residential areas

Variables Mean (SD) U p
Sex
Female Male
Depression 7.09 (3.90) 6.18 (4.18) 65979.50 < .001
Anxiety 7.81 (3.80) 6.49 (3.92) 62131.00 < .001
Stress 9.28 (3.50) 7.96 (4.04) 60719.50 < .001
Internet addiction 29.84 (8.76) 27.53 (10.00) 64958.00 < .001
Self-esteem 16.37 (4.77) 17.09 (4.72) 70244.50 .030
The family history of psychotic diseases
Yes No
Depression 8.25 (3.83) 6.57 (4.05) 13972.50 .001
Anxiety 9.12 (4.09) 7.08 (3.86) 13717.50 .001
Stress 10.13 (3.30) 8.58 (3.83) 14321.00 .002
Internet addiction 32.48 (8.54) 28.53 (9.41) 14037.00 .001
Self-esteem 15.23 (4.38) 16.80 (4.77) 14494.00 .003
The family history of alcohol addiction
Yes No
Depression 7.51 (3.71) 6.60 (4.08) 20919.00 .019
Anxiety 8.49 (4.50) 7.09 (3.82) 20491.50 .010
Stress 10.19 (4.04) 8.54 (3.76) 18749.50 < .001
Internet addiction 30.26 (9.14) 28.65 (9.43) 22183.00 .101
Self-esteem 15.83 (4.98) 16.78 (4.73) 22185.00 .101
Residential areas
Rural areas Urban areas
Depression 7.39 (3.84) 6.59 (4.07) 25207.00 .052
Anxiety 7.38 (3.82) 7.19 (3.92) 27854.00 .560
Stress 8.96 (3.35) 8.65 (3.86) 27037.00 .316
Internet addiction 26.30 (5.85) 28.65 (9.44) 25624.50 .085
Self-esteem 15.65 (4.28) 16.82 (4.80) 24962.50 .039
Open in a new tab

Model specification

Hair et al. [23] suggest that VIF values below 3 indicate no multicollinearity issues. The results showed that all constructs had VIF values ranging from 1.11 to 2.22, confirming that collinearity among latent variables was not problematic. Model fit was evaluated using the SRMR. Consistent with the recommended threshold of 0.09 [19], the obtained SRMR value of 0.07 indicates an adequate fit between the proposed model and the observed data. Regarding the coefficient of determination, the R² values of 0.75, 0.50, and 0.25 are commonly interpreted as substantial, moderate, and weak, respectively [22]. The R² values observed in the present study ranged from 0.16 to 0.37, reflecting a weak to moderate level of explanatory power. This suggests that the endogenous constructs were explained to a limited but acceptable extent by the predictors included in the model. Predictive relevance was assessed using the Q² statistic, where values greater than 0 indicate that the model has predictive capability for the endogenous constructs. The Q² values obtained in this study were all positive, thereby supporting the predictive relevance of the proposed PLS-SEM model [76]. The effect size was examined using f², with values of 0.02, 0.15, and 0.35 representing weak, moderate, and strong effects, respectively [21]. The results revealed moderate effect sizes for the relationships between self-esteem and depression (f² = 0.48), anxiety (f² = 0.40), and stress (f² = 0.19). In contrast, the effects of depression, anxiety, and stress on internet addiction were weak, with f² values ranging from 0.02 to 0.06. Bootstrap confidence intervals can be employed to examine whether the HTMT ratios differ significantly from 1.0 [26] or from more conservative cutoff values, such as 0.90 or 0.80 [20]. The results indicated that all HTMT values among the constructs were below the recommended threshold of 0.80, providing evidence of adequate discriminant validity.

Although the AVE values for these constructs were below the recommended threshold of 0.50, the CR exceeded 0.70 for all constructs. According to Fornell and Larcker [16], constructs with AVE values may still be considered adequate if CR is satisfactory, indicating sufficient internal consistency despite lower variance extracted. For example, previous validation studies of the DASS-21 in Vietnamese and other Asian adolescents and young adults’ populations have similarly reported lower AVE values for the stress subscale, while retaining acceptable composite reliability and theoretical coherence (i.e., Jiang et al. [31], Lan, [38]). Such findings have been attributed to cultural differences in stress expression among adolescents, where stress-related symptoms tend to overlap with anxiety and somatic complaints. The loadings varied from 0.28 to 0.88 and were all statistically significant (p < .001). Table 3 demonstrates the reliability and validity of the endogenous latent variables. The AVEs for the constructs range from 0.39 to 0.48 and were slightly below the recommended threshold of 0.50, the composite reliability values exceeded 0.70, suggesting acceptable convergent validity according to Fornell and Larcker [16]. Cronbach’s alpha coefficients vary from 0.74 to 0.88. The CR coefficients ranged from 0.82 to 0.90, indicating substantial shared variance among the indicators.

Table 3.

Average Variance Extracted (AVE), Construct Reliability (CR), and Cronbach’s alpha among Depression, Anxiety, Stress, Self-esteem, and Internet Addiction (N = 789)

AVE CR (rho_c) Cronbach’s
alpha
Depression 0.47 0.86 0.82
Anxiety 0.48 0.86 0.81
Stress 0.39 0.82 0.74
Self-esteem 0.43 0.87 0.83
Internet addiction 0.43 0.90 0.88
Open in a new tab

Findings for the structural model proposed in the present study

The findings of the study revealed that self-esteem was negatively associated with depression (β = -0.57, 95% CI [-0.62; -0.52], p < .001), anxiety (β = -0.54, 95% CI [-0.59; -0.49], p < .001), stress (β = -0.40; 95% CI [-0.47; -0.34], p < .001), and internet addiction (β = -0.29, 95% CI [-0.36; -0.23], p < .001), thereby supporting Hypothesis 1a, 1b, 1c, and 1d.

Depression (β = 0.18, 95% CI [0.06; 0.29], p = .002), anxiety (β = 0.21, 95% CI [0.10; 0.33], p < .001), and stress (β = 0.31, 95% CI [0.22; 0.40], p < .001) was positively associated with internet addiction, providing empirical support for Hypotheses 2a, 2b, and 2c.

Furthermore, the study identified significant indirect effects of self-esteem on internet addiction through mental health problems. The total effect of self-esteem on internet addiction was significant (β = -0.29, 95% CI [-0.36; -0.23], p < .001). The indirect effects through depression (β = -0.10, 95% CI [-0.17; -0.04], p = .003), anxiety (β = -0.12, 95% CI [-0.18; -0.05], p < .001), and stress (β = -0.12, 95% CI [-0.17; -0.09], p < .001) were all significant, supporting Hypotheses 3a, 3b, and 3c (Table 4).

Table 4.

The Mediating Effect of Mental Health Problems in the Relationship between Self-esteem and Internet Addiction (N = 789)

Hypothesis Path β Confidence Intervals 95% t p
Direct effect
 H1a Self-esteem → Depression -0.57 [ -0.62; -0.52] 22.50 < .001
 H1b Self-esteem → Anxiety -0.54 [ -0.59; -0.49] 19.81 < .001
 H1c Self-esteem → Stress -0.40 [-0.47; -0.34] 12.83 < .001
 H1d Self-esteem → Internet Addiction -0.29 [-0.36; -0.23] 8.60 < .001
 H2a Depression → Internet Addiction 0.18 [0.06; 0.29] 3.06 .002
 H2b Anxiety → Internet Addiction 0.21 [0.10; 0.33] 3.72 < .001
 H2c Stress → Internet Addiction 0.31 [0.22; 0.40] 6.71 < .001
Indirect effect
 H3a Self-esteem → Depression → Internet Addiction -0.10 [-0.17, -0.04] 3.00 .003
 H3b Self-esteem → Anxiety → Internet Addiction -0.12 [-0.18, -0.05] 3.63 < .001
 H3c Self-esteem → Stress → Internet Addiction -0.12 [-0.17, -0.09] 5.94 < .001
Open in a new tab

Figure 1 displays the path coefficients, which elucidate the correlations among the variables. Although the total effect of self-esteem on internet addiction was significant (β = -0.29, 95% CI [-0.36; -0.23], p < .001), the direct effect became non-significant after including the mediators (β = 0.05, p = .180). This indicates that the relationship between self-esteem and internet addiction was fully mediated by depression, anxiety, and stress.

Fig. 1.

Fig. 1

Open in a new tab

The mediating model of mental health problems on self-esteem and internet addiction

Discussion

The present study aimed to investigate the interrelationships among mental health problems (i.e., depression, anxiety, and stress), self-esteem, and internet addiction, as well as the roles of mental health problems in the association between self-esteem and internet addiction. These findings were specific to Vietnamese high school students and were interpreted within Vietnam’s sociocultural context. Rapid digitalization, high academic pressure, and strong family- and school-oriented expectations may intensify the interplay between self-esteem, mental health problems, and internet addiction in this population [39, 53, 54, 59, 80].

The study found that self-esteem was negatively associated with depression, anxiety, stress, and internet addiction, indicating that individuals with lower self-esteem are more likely to experience mental health problems. Our findings are consistent with a substantial body of literature identifying low self-esteem as a significant risk factor for the development of various mental health problems [5, 6, 25, 27, 47, 54, 61, 69, 71]. Individuals with persistently low self-esteem are prone to negative self-evaluation and rumination, cognitive patterns that progressively increase their susceptibility to depression and anxiety [28]. Low self-esteem reinforces these maladaptive thought patterns, leading to persistent negative mood states and depressive symptoms [5, 6]. In contrast, individuals with high self-esteem are more likely to appraise challenging situations as manageable or even growth-promoting, thereby experiencing less anxiety [25, 54].

Depression, anxiety, and stress were positively associated with internet addiction, aligning with substantial empirical evidence indicating that individuals experiencing mental health problems are more vulnerable to problematic internet use [30, 68, 82, 86, 88, 90]. This relationship can be explained by the I-PACE theory [7], showing the relationship between students’ personal characteristics and cognitive, emotional, and executive processes [7]. The internet offers immediate gratification and a sense of control, which may temporarily alleviate depressive feelings but can lead to compulsive online use over time [86, 88].

The social support and coping models suggest that individuals with mental health problems may use the internet as a maladaptive coping strategy to regulate negative emotions or to escape from real-world distress [7, 77]. The internet provides an easily accessible, socially less threatening environment that can temporarily alleviate anxiety. However, this coping mechanism may lead to habitual or compulsive use, thereby increasing the risk of addiction [34]. For example, Yao and Zhong , [87] found that anxiety symptoms significantly predicted problematic internet use among Chinese adolescents. Similarly, Weinstein and Lejoyeux , [82] emphasized that anxiety disorders, particularly social anxiety, are common among individuals with internet addiction.

The present study found that mental health problems partially mediated the relationship between self-esteem and internet addiction. Lower self-esteem increases the risk of depression, which in turn is associated with higher levels of addictive internet use. This result is consistent with previous evidence that self-esteem serves as a critical protective factor against the development of depression and behavioral addictions [69]. Individuals with low self-esteem often lack adaptive coping strategies and are more likely to feel helpless in the face of psychological pressures, which increases their vulnerability to depressive symptoms [61]. Anxiety serves as a mediating role in the relationship between self-esteem and internet addiction. This mechanism can be explained by the “compensatory internet use” model, which suggests that individuals with low self-esteem and heightened anxiety often resort to the internet as a coping tool to alleviate stress, avoid anxiety-provoking situations, or seek a sense of safety in virtual environments [35]. This study contributes to expanding the literature by clarifying the mechanisms through which mental health problems influence internet addiction and highlights the importance of integrated interventions aimed at enhancing self-esteem and reducing mental health problems for both the prevention and treatment of internet addiction.

Findings from our study additionally showed that students characterized by low happiness, having negative feelings toward school, and having a family history of psychotic illness, were at heightened risk of internet addiction and low self-esteem. These findings align with those of previous studies [18, 39, 43]. In contrast, family history of alcohol addiction and residential area were not significantly associated with internet addiction in the present sample. Notably, adolescents who engage in problematic internet use frequently report diminished happiness and life satisfaction [13, 37, 73], with research from Vietnam, South Korea, and China suggesting that reduced happiness functions both as a contributing factor to and a consequence of internet addiction [63, 91]. Experiences of pessimism and hopelessness may be intensified when individuals perceive the learning environment as unsafe, lacking support, or excessively demanding [52]. Moreover, adolescents with a family history of psychotic diseases are at higher risk of developing mental health problems and internet addiction [18, 39, 43]. Investigating these background variables may help understand the multifaceted interactions among environmental, educational, emotional, and familial influences on adolescent mental health. Such insights are vital for shaping effective interventions and educational policies that foster student well-being.

Limitation

Several limitations of the present study should be acknowledged. First, the predominantly cross-sectional design restricts the ability to draw causal inferences regarding the relationships among self-esteem, mental health problems, and internet addiction. Longitudinal studies are needed to establish the direction of these associations. Second, the reliance on self-report measures may introduce biases, such as social desirability and recall errors, particularly when assessing sensitive questions, including depression, anxiety, stress, and internet addiction. Future studies could incorporate qualitative studies, multiple data sources, and objective measures to mitigate these biases. It should consider incorporating in-depth interviews and observations to enhance data accuracy and control potential errors. This would provide more comprehensive and detailed insights into self-esteem, mental health problems, and internet addiction among high school students. Third, the use of convenience sampling from two high schools limits the generalizability of the findings to broader adolescent populations. Expanding the sample to include students from diverse schools and geographic regions would strengthen the external validity of future research. Fourth, cultural and contextual differences, such as variations in educational systems and family environments, may limit the generalizability of the findings.

Finally, relevant demographic (e.g., age, sex, and grade level) or important psychosocial variables (e.g., happiness and school-related feelings collected in the present study) were not controlled in the PLS-SEM. Therefore, the relationships between mental health problems, self-esteem, and internet addiction might be changed if these variables were controlled. However, we believe that the PLS-SEM findings without controlling demographic and psychosocial variables can provide initial and clear information on the associations between mental health problems, self-esteem, and internet addiction; and satisfy the principle of parsimony in statistics.

Implications

The results of this study carry the following implications. From a preventive perspective, interventions aimed at enhancing self-esteem and fostering adaptive coping strategies may reduce the risk of internet addiction, particularly among adolescents experiencing elevated levels of mental health problems. School-based programs that incorporate psychoeducation, stress management, and social-emotional learning could strengthen psychological resilience and promote healthier online behaviors. Clinically, the early identification of adolescents with low self-esteem and co-occurring mental health problems may enable more timely and targeted interventions. Moreover, future researchers may consider adopting qualitative methods (e.g., Online Photovoice [OPV] and Online Interpretative Phenomenological [OIPA] with Community-Based Participatory Research [CBPR] approach) and the constructs of self-compassion, perfectionistic self-presentation, and entitlement to conduct research on the associations between mental health, self-esteem, and internet addiction. Specifically, OPV and/or OIPA can collaborate with communities from a CBPR perspective to explore the mediating role of depression, anxiety, and stress in the relationship between self-esteem and internet addiction among adolescents [11, 14, 75]. With the use of OPV and OIPA, participants will have opportunities to express their own experience and supplement the findings from traditional quantitative methods.

Conclusion

Our findings explored the relationships between depression, anxiety, stress, internet addiction, and self-esteem, which are particularly relevant to Vietnamese adolescents. Notably, depression, anxiety, and stress were positively related to internet addiction, while self-esteem was negatively associated with depression, anxiety, and stress. The relationship between self-esteem and internet addiction was partially mediated by depression, anxiety, and stress. These results provide valuable implications for the design and implementation of targeted intervention strategies. Fostering self-esteem and addressing underlying psychological difficulties may help reduce the risk of internet addiction among Vietnamese high school students. Notably, these findings highlight the need for culturally and contextually tailored mental health promotion and internet-use interventions at the national level.

Acknowledgements

The authors sincerely thank the students who participated in the survey and their guardians for providing informed consent. We are also grateful to the principals of the two participating high schools in Can Tho City, Vietnam, for their support. Special appreciation is extended to Can Tho University of Medicine and Pharmacy and National Cheng Kung University for their facilitation of this research. Additionally, we acknowledge that ChatGPT is used solely to check grammar, wording, and sentence structure. It does not contribute to study design, data collection, analysis, interpretation of results, or intellectual content of the manuscript.

Authors’ contributions

DTN, TTTN and V-LT-C jointly generated the research idea and developed the study design. DTN, TTTN, and NHP coordinated the surveys and field data collection. Data analysis was done by DTN, TTTN and V-LT-C. DTN and TTTN was guided by C-YL and V-LT-C to write the first draft. TDH, HNN, CSHH, and KL helped with methodology and interpretation of results. All authors provided critical comments and contributed to the final version of the manuscript. All authors read and approved the final manuscript.

Funding

The current investigation did not obtain any external financial support.

Data availability

The data that support the findings of this study are available on request from the corresponding author.

Declarations

Ethics approval and consent to participate

Approval for the study was granted by the Ethics Review Board of the University of Social Sciences and Humanities, Viet Nam National University Ho Chi Minh City, Vietnam (Approval ERB- 31/2024 No. 1431/GXN-XHNV-ĐN&QLKH), in compliance with the ethical guidelines of the Declaration of Helsinki (World Medical Association, 2013) and the ethical standards of the American Psychological Association (APA). All participants were fully informed about the study procedures, risks, and benefits, and provided written informed consent prior to participation. Participant confidentiality was ensured by collecting data anonymously and limiting access exclusively to the main researcher.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Thao Thanh Thi Nguyen, Email: nttthao@ctump.edu.vn.

Chung-Ying Lin, Email: cylin36933@gmail.com.

References

  • 1.Ahorsu DK. Pathways to social media addiction: Examining its prevalence, and predictive factors among Ghanaian youths. J Social Media Res. 2024;1(1):47–59. 10.29329/jsomer.9. [Google Scholar]
  • 2.Angane AY, Kadam KS, Ghorpade GS, Unnithan VB. Unraveling the net of self-esteem, stress, and coping skills in the era of Internet addiction. Annals Indian Psychiatry. 2020;4(1):70–5. 10.4103/aip.aip_80_19. [Google Scholar]
  • 3.Angulo PG, Mendoza JS, Contreras EC, Martínez JUC, Hernández CBE, Hernández MEC. The meaning of life as a mediator between self-esteem and internet addiction in adolescents. Enfermería Global. 2021;20(4):519–31. 10.6018/eglobal.482691. [Google Scholar]
  • 4.Arab M, Rafiei H, Safarizadeh MH, Ahmadi JS, Safarizadeh MM. Stress, anxiety and depression among medical university students and its relationship with their level of happiness. J Nurs Health Sci. 2016;5:44–7. 10.9790/1959-05164447. [Google Scholar]
  • 5.Babore A, Trumello C, Candelori C, Paciello M, Cerniglia L. Depressive symptoms, self-esteem and perceived parent-child relationship in early adolescence. Front Psychol. 2016;7(982):1–7. 10.3389/fpsyg.2016.00982. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Batra P. Self-esteem and Internet addiction among late adolescents and young adults. Indian J Health Wellbeing. 2020;11(7–9):320–3. [Google Scholar]
  • 7.Brand M, Wegmann E, Stark R, Müller A, Wölfling K, Robbins TW, Potenza MN. The Interaction of Person-Affect-Cognition-Execution (I-PACE) model for addictive behaviors: Update, generalization to addictive behaviors beyond internet-use disorders, and specification of the process character of addictive behaviors. Neurosci Biobehavioral Reviews. 2019;104:1–10. 10.1016/j.neubiorev.2019.06.032. [DOI] [PubMed] [Google Scholar]
  • 8.Cao LX, Le CM, Tran DTT, Nguyen VHA, Tran-Chi V-L. Prevalence and factors associated with depression among high school students in Ho Chi Minh City, Vietnam. Natl J Community Med. 2024;15(7):526–32. 10.55489/njcm.150720243843. [Google Scholar]
  • 9.Cao H, Sun Y, Wan Y, Hao J, Tao F. Problematic internet use in Chinese adolescents and its relation to psychosomatic symptoms and life satisfaction. BMC Public Health. 2011;11(802):1–8. 10.1186/1471-2458-11-802. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Chen H-C, Wang J-Y, Lin Y-L, Yang S-Y. Association of internet addiction with family functionality, depression, self-efficacy and self-esteem among early adolescents. Int J Environ Res Public Health. 2020;17(23):1–13. 10.3390/ijerph17238820. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Dari T, Fox C, Laux JM, Speedlin Gonzalez S. The Development and Validation of the Community-Based Participatory Research Knowledge Self-Assessment Scale (CBPR-KSAS): A Rasch Analysis. Meas Evaluation Couns Dev. 2023;56(1):64–79. 10.1080/07481756.2022.2034478. [Google Scholar]
  • 12.Dash G, Paul J. CB-SEM vs PLS-SEM methods for research in social sciences and technology forecasting. Technol Forecast Soc Chang. 2021;173:1–11. 10.1016/j.techfore.2021.121092. [Google Scholar]
  • 13.Diener E, Tay L. A scientific review of the remarkable benefits of happiness for successful and healthy living. Happiness: Transforming Dev Landsc. 2017;6:90–117. [Google Scholar]
  • 14.Doyumğaç İ, Tanhan A, Kıymaz MS. Understanding the most important facilitators and barriers for online education during COVID-19 through online photovoice methodology. Int J High Educ. 2021;10(1):166–90. 10.5430/ijhe.v10n1p166. [Google Scholar]
  • 15.Erskine HE, Maravilla JC, Wado YD, Wahdi AE, Loi VM, Fine SL, Scott JG. Prevalence of adolescent mental disorders in Kenya, Indonesia, and Viet Nam measured by the National Adolescent Mental Health Surveys (NAMHS): A multi-national cross-sectional study. Lancet. 2024;403:1671–80. 10.1016/S0140-6736(23)02641-7. [DOI] [PubMed] [Google Scholar]
  • 16.Fornell C, Larcker DF. Evaluating structural equation models with unobservable variables and measurement error. J Mark Res. 1981;18(1):39–50. 10.1177/002224378101800104. [Google Scholar]
  • 17.Gan WY, Chin WL, Huang SW, Tung SEH, Lee LJ, Poon WC, Lin CY. Association between mental distress and weight-related self-stigma via problematic social media and smartphone use among Malaysian university students: An application of the interaction of person-affect-cognition-execution (I-PACE) model. Int J Mental Health Promotion. 2025;27(3):319–31. 10.32604/ijmhp.2025.060049. [Google Scholar]
  • 18.Ghimire R, Mishra A, T., Sharma P. Stress, anxiety and depression among adolescents of schools in Godawari municipality, Lalitpur. Kathmandu Univ Med J. 2024;88(4):426–32. [PubMed] [Google Scholar]
  • 19.Hair J, Black W, Babin B, Anderson R. (2018). Multivariate data analysis (8th ed.). Cengage.
  • 20.Hair JF, Hollingsworth CL, Randolph AB, Chong AYL. An updated and expanded assessment of PLS-SEM in information systems research. Industrial Manage Data Syst. 2017;117(3):442–58. 10.1108/IMDS-04-2016-0130. [Google Scholar]
  • 21.Hair JF, Hult GTM, Ringle CM, Sarstedt M. A primer on partial least squares structural equation modeling (PLS-SEM). Sage; 2021.
  • 22.Hair JF, Ringle CM, Sarstedt M. PLS-SEM: Indeed a silver bullet. J Mark Theory Pract. 2011;19(2):139–52. 10.2753/MTP1069-6679190202. [Google Scholar]
  • 23.Hair JF, Risher JJ, Sarstedt M, Ringle CM. When to use and how to report the results of PLS-SEM. Eur Bus Rev. 2019;31(1):2–24. 10.1108/EBR-11-2018-0203. [Google Scholar]
  • 24.Hair J. (2009). Multivariate Data Analysis. Retrieved from https://digitalcommons.kennesaw.edu/facpubs/2925
  • 25.Henriksen IO, Ranøyen I, Indredavik MS, Stenseng F. The role of self-esteem in the development of psychiatric problems: a three-year prospective study in a clinical sample of adolescents. Child Adolesc Psychiatry Mental Health. 2017;11(68):1–9. 10.1186/s13034-017-0207-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Henseler J, Ringle CM, Sarstedt M. A new criterion for assessing discriminant validity in variance-based structural equation modeling. J Acad Mark Sci. 2015;43(1):115–35. 10.1007/s11747-0. [Google Scholar]
  • 27.Hesapcioglu ST, Yesilova H. Internet addiction prevalence in youths and its relation with depressive symptoms, self-esteem, and bullying. Alpha Psychiatry. 2020;21(5):483–90. 10.5455/apd.72881. [Google Scholar]
  • 28.Huang D. (2021). The effect of low self-esteem on development of depression: A meta-analysis. In Proceedings of the 2021 6th International Conference on Modern Management and Education Technology (MMET 2021) (pp. 388–393). Atlantis Press. 10.2991/assehr.k.211011.070
  • 29.Huang YT, Huang PC, Hou WL, Aljaberi MA, Gan WY, Tung SEH, Lin CY. Using moderated mediation modelling and the Interaction of Person-Affect-Cognition-Execution Model to explore relationships between psychological distress, specific addictive behaviors, and quality of life across Taiwan, Malaysia, Hong Kong, and China. Appl Res Qual Life. 2025;20(4):1759–82. 10.1007/s11482-025-10495-1. [Google Scholar]
  • 30.Ismail A. The relationship between internet addiction and stress, anxiety, and depression in adolescents in Saudi Arabia. J Pediatr Nurs. 2024;77:111–6. 10.1016/j.pedn.2024.03.007. [DOI] [PubMed] [Google Scholar]
  • 31.Jiang J, Chen J, Lin Z, Tang X, Hu Z. Validation and psychometric properties of the depression anxiety stress scale for youth in Chinese adolescents. Front Psychol. 2024;15:1–8. 10.3389/fpsyg.2024.1466426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Kantar A, Kök M. Masculinity and its association with English-speaking anxiety, self-compassion, and perfectionistic self-presentation. Gend Issues. 2025;42(4):291–310. 10.1007/s12147-025-09374-8. [Google Scholar]
  • 33.Kantar A, Yalçın İ, Kocabıyık OO, Barry C. Joy of Missing Out (JOMO) and its role in reducing social media addiction: The serial mediating role of loneliness and psychological distress. Psychol Rep. 2025;0(0):1–23. 10.1177/00332941251378710. [DOI] [PubMed] [Google Scholar]
  • 34.Kardefelt-Winther D. A conceptual and methodological critique of internet addiction research: Towards a model of compensatory internet use. Comput Hum Behav. 2014a;31:351–4. 10.1016/j.chb.2013.10.059. [Google Scholar]
  • 35.Kardefelt-Winther D. The moderating role of psychosocial well-being on the relationship between escapism and excessive online gaming. Comput Hum Behav. 2014b;38:68–74. 10.1016/j.chb.2014.05.020. [Google Scholar]
  • 36.Kline RB. Principles and Practice of Structural Equation Modeling. Guilford Press; 2023.
  • 37.Ko CH, Yen JY, Yen CF, Chen CS, Chen CC. The association between internet addiction and psychiatric disorder: A review of the literature. Eur Psychiatry. 2012;27(1):1–8. 10.1016/j.eurpsy.2010.04.011. [DOI] [PubMed] [Google Scholar]
  • 38.Lan HTQ, Long NT, Hanh NV. Validation of depression, anxiety and stress scales (DASS-21): Immediate psychological responses of students in the e-learning environment. Int J High Educ. 2020;9(5):125–33. 10.5430/ijhe.v9n5p125. [Google Scholar]
  • 39.Le SQ, Bui DTT, Le DT, Nguyen TTP, Dam NTK. The relationship between depression, anxiety, stress and internet use among high school students. Asean J Psychiatry. 2024;25(5):1–16. 10.54615/2231-7805.47354. [Google Scholar]
  • 40.Le MTH, Tran TD, Holton S, Nguyen HT, Wolfe R, Fisher J. Reliability, convergent validity and factor structure of the DASS-21 in a sample of Vietnamese adolescents. PLoS ONE. 2017;12(7):1–14. 10.1371/journal.pone.0180557. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Lei H, Xiong J, Rao Y, Zhu T, Zhang X. Relationships among self-esteem, depression and self-injury in adolescents: A longitudinal study. Front Public Health. 2024;12:1–9. 10.3389/fpubh.2024.1406283. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Li W, Guo Y, Lai W, et al. Reciprocal relationships between self-esteem, coping styles and anxiety symptoms among adolescents: Between-person and within-person effects. Child Adolesc Psychiatry Mental Health. 2023;17(21):1–16. 10.1186/s13034-023-00564-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Liu L, Tian Y, Fan H, Wang J, Chen C, Liu Z, Geng F, Mo D, Luo X, Wen X, Zhao X, Hao M, Xia L, Liu H. Associations between internet addiction and suicidal ideation in depressed adolescents: the mediating effect of insomnia as well as sex differences. BMC Psychiatry. 2024;24(1):1–12. 10.1186/s12888-024-06357-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Lovibond PF, Lovibond SH. The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behav Res Ther. 1995;33(3):335–43. 10.1016/0005-7967(94)00075-U. [DOI] [PubMed] [Google Scholar]
  • 45.Luu-Thi HT, Nguyen-Thi TT, Mallari EFI, Santos LE, Peltzer K. Trends in suicidal behavior among adolescents in the Philippines (2011 to 2019) and associations with being bullied, alcohol use, and parental involvement. Psychol Health. 2026;1–22. 10.1080/08870446.2026.2632070. [DOI] [PubMed]
  • 46.Mallari EFI, Peltzer K, Santos LE, Luu-Thi HT, Nguyen-Thi TT, Lin K, Suen MW. Active travel to school, psychological distress, health risk behaviors, and social-environmental factors among Filipino adolescents. J Phys Activity Health. 2026;1:1–8. 10.1123/jpah.2025-0791. [DOI] [PubMed] [Google Scholar]
  • 47.Mann M, Hosman CM, Schaalma HP, de Vries NK. Self-esteem in a broad-spectrum approach for mental health promotion. Health Educ Res. 2004;19(4):357–72. 10.1093/her/cyg041. [DOI] [PubMed] [Google Scholar]
  • 48.Mehrzad K, Yazdanpanah F, Arab M, Ghasemi M, Radfar A. Relationship between stress, anxiety, and depression with happiness in students of Bam medical university in 2019. J Adv Pharm Educ Res. 2022;12(2):51–6. 10.51847/dJZ1dCmMK6. [Google Scholar]
  • 49.Minev M, Petrova B, Mineva K, Petkova M, Strebkova R. Self-esteem in adolescents. Trakia J Sci. 2018;16(2):114–8. 10.15547/tjs.2018.02.007. [Google Scholar]
  • 50.Moksnes UK, Reidunsdatter RJ. Self-esteem and mental health in adolescents–level and stability during a school year. Norsk Epidemiologi. 2019;28(1–2):59–67. 10.5324/nje.v28i1-2.3052. [Google Scholar]
  • 51.Moksnes UK, Espnes GA. Self-esteem and emotional health in adolescents–gender and age as potential moderators. Scand J Psychol. 2012;53(6):483–9. 10.1111/sjop.12021. [DOI] [PubMed] [Google Scholar]
  • 52.Mori Y, Tiiri E, Khanal P, Khakurel J, Mishina K, Sourander A. Feeling unsafe at school and associated mental health difficulties among children and adolescents: A systematic review. Children. 2021;8(3):1–18. 10.3390/children8030232. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Nguyen DT, Dedding C, Pham TT, Bunders J. Perspectives of pupils, parents, and teachers on mental health problems among Vietnamese secondary school pupils. BMC Public Health. 2013;13(1):1–10. 10.1186/1471-2458-13-1046. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Nguyen DT, Wright EP, Dedding C, Pham TT, Bunders J. Low self-esteem and its association with anxiety, depression, and suicidal ideation in Vietnamese secondary school students: A cross-sectional study. Front Psychiatry. 2019;10:1–7. 10.3389/fpsyt.2019.00698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Nguyen TTT, Luu-Thi HT, Nguyen DT, Pham NH, Peltzer K. Associations between child labour, child discipline and psychosocial functioning of children and adolescents in Laos. Child Care Health Dev. 2026;52(2):1–16. 10.1111/cch.70260. [DOI] [PubMed] [Google Scholar]
  • 56.Nguyen T, Nguyen N, Pham NH, Tran LC, Tran N, Nguyen DT. Internet addiction and related factors among high school students in Vietnam. Annali di Igiene Med Preventiva e di Comunità. 2026;38(1):1–12. 10.7416/ai.2026.17913. [Google Scholar]
  • 57.Nguyen-Thi DM, Huynh VS, Tran-Chi VL. Loneliness, stress, self-esteem, and deception among adolescents. J Hum Ecol. 2020;70(1–3):118–23. 10.31901/24566608.2020/70.1-3.2593. [Google Scholar]
  • 58.Nguyen-Thi TT, Le HM, Chau TL, Le HT, Pham TT, Tran NT, Nguyen DT. Prevalence of stress and related factors among healthcare students: A cross-sectional study in Can Tho City, Vietnam. Annali di Igiene Med Preventiva e di Comunità. 2024;36(3):292–301. 10.7416/ai.2023.2591. [DOI] [PubMed] [Google Scholar]
  • 59.Nguyen-Thi TT, Nguyen DT, Le HM, Le CM, Hua TD, Nguyen-Hoang BN, Tran-Chi VL. Academic stress and attitudes toward seeking professional psychological help among Vietnamese students: the mediating role of emotional intelligence. Discover Social Sci Health. 2024;4(1):1–18. 10.1007/s44155-024-00110-3. [Google Scholar]
  • 60.Ninh LK. Economic role of education in agriculture: Evidence from rural Vietnam. J Econ Dev. 2021;23(1):47–58. 10.1108/JED-05-2020-0052. [Google Scholar]
  • 61.Orth U, Robins RW. The development of self-esteem. Curr Dir Psychol Sci. 2014;23(5):381–7. 10.1177/0963721414547414. [Google Scholar]
  • 62.Orth U, Robins RW, Meier LL. Disentangling the effects of low self-esteem and stressful events on depression: Findings from three longitudinal studies. J Personal Soc Psychol. 2009;97(2):307–21. 10.1037/a0015645. [DOI] [PubMed] [Google Scholar]
  • 63.Park N, Huebner ES. A cross-cultural study of the levels and correlates of life satisfaction among adolescents. J Cross-Cult Psychol. 2005;36(4):444–56. 10.1177/0022022105275961. [Google Scholar]
  • 64.Phetphum C, Keeratisiroj O, Prajongjeep A. The association between mobile game addiction and mental health problems and learning outcomes among Thai youths classified by gender and education levels. Asian J Social Health Behav. 2023;6(4):196–202. 10.4103/shb.shb_353_23. [Google Scholar]
  • 65.Rosenberg M, Schooler C, Schoenbach C, Rosenberg F. Global self-esteem and specific self-esteem: Different concepts, different outcomes. Am Sociol Rev. 1995;60(1):141–56. 10.2307/2096350. [Google Scholar]
  • 66.Shah SM, Al Dhaheri F, Albanna A, Al Jaberi N, Eissaee A, Alshehhi S, N. A., et al. Self-esteem and other risk factors for depressive symptoms among adolescents in United Arab Emirates. PLoS ONE. 2020;15(1):1–16. 10.1371/journal.pone.0227483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 67.Siaw YL, Chua KH, Ruckwongpatr K, Tung SEH, Gan WY, Poon WC, Lin CY. Psychometric properties for the Malay version of the assessment of criteria for specific internet-use disorders among young adults in Malaysia. Asian J Social Health Behav. 2025;8(3):97–106. 10.4103/shb.shb_171_24. [Google Scholar]
  • 68.Soriano-Molina E, Limiñana-Gras RM, Patró-Hernández RM, Rubio-Aparicio M. The association between internet addiction and adolescents’ mental health: A meta-analytic review. Behav Sci. 2025;15(2):1–34. 10.3390/bs15020116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69.Sowislo JF, Orth U. Does low self-esteem predict depression and anxiety? A meta-analysis of longitudinal studies. Psychol Bull. 2013;139(1):213–40. 10.1037/a0028931. [DOI] [PubMed] [Google Scholar]
  • 70.Spada MM. An overview of problematic internet use. Addict Behav. 2014;39(1):3–6. 10.1016/j.addbeh.2013.09.007. [DOI] [PubMed] [Google Scholar]
  • 71.Steiger AE, Allemand M, Robins RW, Fend HA. Low and decreasing self-esteem during adolescence predict adult depression two decades later. J Personal Soc Psychol. 2014;106(2):325–38. 10.1037/a0035133. [DOI] [PubMed] [Google Scholar]
  • 72.Steinfield C, Ellison NB, Lampe C. Social capital, self-esteem, and use of online social network sites: A longitudinal analysis. J Appl Dev Psychol. 2008;29(6):434–45. 10.1016/j.appdev.2008.07.002. [Google Scholar]
  • 73.Suldo SM, Huebner ES. Is extremely high life satisfaction during adolescence advantageous? Soc Indic Res. 2006;78(2):179–203. 10.1007/s11205-005-8208-2. [Google Scholar]
  • 74.Szkody E, McKinney C. Indirect effects of social support on psychological health through self-esteem in emerging adulthood. J Fam Issues. 2019;40(17):2439–55. 10.1177/0192513X19859612. [Google Scholar]
  • 75.Tanhan A, Strack RW. Online photovoice to explore and advocate for Muslim biopsychosocial spiritual wellbeing and issues: Ecological systems theory and ally development. Curr Psychol. 2020;39(6):2010–25. 10.1007/s12144-020-00692-6. [Google Scholar]
  • 76.Taylor T, Geldenhuys S. (2019). Using partial least squares to measure tourism students’ satisfaction with work-integrated learning. In S. Senay, editor, Tourism-perspectives and practices (pp. 1–20). IntechOpen. 10.5772/intechopen.8204
  • 77.Taylor SE, Stanton AL. Coping resources, coping processes, and mental health. Ann Rev Clin Psychol. 2007;3(1):377–401. 10.1146/annurev.clinpsy.3.022806.091520. [DOI] [PubMed] [Google Scholar]
  • 78.Touloupis T, Sofologi M, Tachmatzidis D. Pattern of Facebook use by university students during the COVID-19 pandemic: Relations with loneliness and resilience. Social Netw Anal Min. 2023;13(1):1–16. 10.1007/s13278-023-01073-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 79.Tran TD, Tran T, Fisher J. Validation of the depression anxiety stress scales (DASS) 21 as a screening instrument for depression and anxiety in a rural community-based cohort of northern Vietnamese women. BMC Psychiatry. 2013;13(24):1–7. 10.1186/1471-244X-13-24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 80.Tran BX, Huong LT, Hinh ND, Nguyen LH, Le BN, Nong VM, Tho TD, Latkin C, Zhang MWB, Ho RC. A study on the influence of internet addiction and online interpersonal influences on health-related quality of life in young Vietnamese. BMC Public Health. 2017;17(138):1–8. 10.1186/s12889-016-3983-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 81.Tran BX, Mai HT, Nguyen LH, Nguyen CT, Latkin CA, Zhang MW, Ho RC. Vietnamese validation of the short version of internet addiction test. Addict Behaviours Rep. 2017;6:45–50. 10.1016/j.abrep.2017.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 82.Weinstein A, Lejoyeux M. Internet addiction or excessive internet use. Am J Drug Alcohol Abus. 2010;36(5):277–83. 10.3109/00952990.2010.491880. [DOI] [PubMed] [Google Scholar]
  • 83.Wiechert S, van Bockstaele B, Vertregt M, van Marwijk L, Maric M. Explicit and implicit self-esteem and their associations with symptoms of anxiety and depression in children and adolescents. Eur J Dev Psychol. 2023;20(5):823–38. 10.1080/17405629.2023.2216447. [Google Scholar]
  • 84.Word Health Organization. (2023). Adolescent Health. https://www.who.int/health-topics/adolescent-health#tab=tab_1
  • 85.World Health Organization. (2021). Mental health of adolescents. Retrieved 17 November, 2021, from https://www.who.int/news-room/fact-sheets/detail/adolescent-mental-health
  • 86.Wu J, Wu X, Tarimo CS, Zhao W, Xia Q, Xu X, Yang Y. Network analysis of internet addiction and depression among Chinese adolescents. J Affect Disord. 2025;378:119–27. 10.1016/j.jad.2024.11.031. [DOI] [PubMed] [Google Scholar]
  • 87.Yao MZ, Zhong ZJ. Loneliness, social contacts and internet addiction: A cross-lagged panel study. Comput Hum Behav. 2014;30:164–70. 10.1016/j.chb.2013.08.007. [Google Scholar]
  • 88.Ye XL, Zhang W, Zhao FF. Depression and internet addiction among adolescents: A meta-analysis. Psychiatry Res. 2023;326:1–11. 10.1016/j.psychres.2023.115311. [DOI] [PubMed] [Google Scholar]
  • 89.Young KS. Psychology of computer use: XL. Addictive use of the internet: A case that breaks the stereotype. Psychol Rep. 1996;79(3):899–902. 10.2466/pr0.1996.79.3.899. [DOI] [PubMed] [Google Scholar]
  • 90.Zhang X, Li C. Predictors of adolescents’ psychological distress and internet addiction: The role of interpersonal stress and school burnout. J Child Fam stud. 2024;33:1070–82. 10.1007/s10826-023-02635-8. [Google Scholar]
  • 91.Zhou N, Cao H, Li X, Zhang J, Yao Y, Geng X, Fang X. Internet addiction, problematic internet use, nonproblematic internet use among Chinese adolescents: Individual, parental, peer, and sociodemographic correlates. Psychol Addict Behav. 2018;32(3):365–72. 10.1037/adb0000358. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author.

Articles from BMC Public Health are provided here courtesy of BMC

RESOURCES