Peer relationship, family factors, and mental health in rural children: a network analysis

The transition from childhood to early adolescence is a critical developmental period, particularly in rural areas where unique factors play a significant role. This study examined the interplay of peer and family factors on the mental health, including anxiety and depression, of 694 rural Chinese children in the transition to adolescence (aged 10-14). This study employed a quantitative design, analyzing data from 16 rural schools across seven provinces using Mixed Graphical Model (MGM) and Gaussian Graphical Model (GGM) network analyses to explore the relationships between variables related to family factors, peer relationships, anxiety, depression, and behavioral problems. Network analysis revealed that depression was a central node, and peer relationships had a greater impact on the mental health of these children than family factors. There was a significant correlation between depression and anxiety (weight = 0.55). Peer relationships were negatively associated with depression (weight = -0.31) and positively associated with father involvement. In contrast, the network shows that the experience of being left behind was not directly associated with mental health outcomes. Further analysis of the depression network revealed that negative mood was a central node. This study highlights that peer relationships have a far stronger connection to the mental health of rural Chinese children than family factors. Depression, and specifically negative mood, was central to the network, emphasizing the need for interventions that focus on improving peer interactions and addressing core depressive symptoms.

1 Introduction

The transition from childhood to early adolescence is a critical developmental period marked by significant emotional, social, and psychological changes. This stage is foundational, often establishing mental health patterns that persist into adulthood (1). For children in rural China, this transition is embedded within a context of distinctive socio-cultural and economic factors, creating developmental trajectories that differ markedly from those of their urban peers. A growing body of evidence indicates that rural children may experience higher rates of mental health challenges, including depression and anxiety, compared to their urban counterparts (2). These disparities are frequently linked to unique contextual factors, such as limited access to mental health resources, distinct cultural norms, and specific family dynamics shaped by widespread parental migration (3).

Within this context, family dynamics have long been recognized as a pivotal determinant of psychological well-being (4). Positive parental relationships, characterized by warmth, effective communication, and consistent involvement, are robustly associated with reduced risks of internalizing disorders in children. However, the common phenomenon of parental migration for work in rural China introduces significant challenges to these family dynamics (5). While much literature has focused on the “left-behind” experience, linking it to higher levels of depression and anxiety (6), it is crucial to understand this as one factor within a broader familial system. For instance, the role of father involvement has emerged as a critical factor, with studies indicating that positive paternal engagement can significantly buffer against familial stressors and promote social adjustment, even in the context of migration (7).

Concurrently with these family influences, early adolescence is characterized by a profound developmental shift toward peer-centric social worlds (8). Peer relationships become a primary context for social and emotional development, acting as key conduits for social support, identity formation, and behavioral norms (9). The quality of these interactions has a direct and significant association with mental health; positive peer acceptance and support are linked to better psychological outcomes, whereas peer rejection and victimization can exacerbate distress (10). In the specific context of rural China, these peer networks may play an especially crucial buffering role. Research suggests that supportive peer relationships can mitigate the negative impacts of family adversity, such as the distress associated with parental absence (11).

Despite the well-documented importance of both family and peer domains, a significant gap persists in the literature. The majority of existing studies tend to examine either family or peer influences in isolation, presupposing one or the other as the primary driver of mental health outcomes. This siloed approach fails to capture the complex interplay and, critically, the relative importance of these two social worlds as they co-exist in a child’s daily life. It is particularly unclear how these systems interact for rural children, who must simultaneously navigate factors like parental absence while relying heavily on their immediate, and often highly stable, school-based peer groups. This study seeks to address this gap by conceptualizing these influences not as independent predictors, but as an interconnected system.

To explore this complex system, we employ a network analysis approach. Unlike traditional regression models that focus on linear, unidirectional associations, network analysis provides a holistic, data-driven visualization of the entire web of reciprocal relationships (12). This method allows us to quantify the intricate connections among multiple psychosocial variables simultaneously, identifying which factors are most central and how they directly and indirectly influence one another. Network analysis has emerged as a powerful tool in adolescent mental health research. For instance, Sánchez Hernández et al. (2023) (13) applied network analysis to explore the comorbidity of anxiety and depression symptoms in Spanish children, identifying ‘worry’ and ‘fatigue’ as key bridge symptoms connecting the two disorders. Similarly, Long et al. (2020) (14) employed this approach to investigate associations between mental health and peer relationships, demonstrating that while depression was linked to perceived social isolation, conduct disorders were more related to having antisocial peers. These studies demonstrate the utility of network analysis in understanding complex psychosocial dynamics.

The current study, therefore, utilizes network analysis to examine the intricate system of peer relationships, family factors (including parental relationships and father involvement), and mental health outcomes (depression, anxiety, and behavioral problems) in a large sample of rural Chinese children. Grounded in social development theory, which posits the increasing salience of peers during adolescence (15), we hypothesized that peer relationships would emerge as a highly central component of this psychological network, potentially exerting a stronger influence than family factors. Furthermore, we expected to observe strong comorbidity between depression and anxiety. We also anticipated that the left-behind experience, while an important contextual variable, might not show a direct connection to mental health outcomes, but rather be mediated by more proximal factors within the network, such as the quality of peer and family relationships.

2 Method

2.1 Data source

This study employed a quantitative cross-sectional design, analyzing secondary data from the ‘Elementary School Students’ Living Conditions Survey’ conducted in 2021 (16). A multi-stage sampling strategy was utilized by the original researchers to ensure a representative sample of rural Chinese students. Initially, 16 rural schools were selected from seven provinces across diverse geographical and socioeconomic regions of China: Anhui, Gansu, Guangdong, Heilongjiang, Hubei, Hunan, and Sichuan.

All students from grades one through six at these schools were invited to participate. From 3,025 distributed questionnaires, 2,498 valid responses were obtained (an 82.28% effective response rate) after excluding those with significant missing data or careless responding. For the purpose of the present analysis, the sample was necessarily restricted to participants in grades five and six. This selection criterion was dictated by the original data collection protocol, in which key psychological instruments were administered only to the older students (grades 5-6) due to the age appropriateness of the scale and to reduce response burden. Our final analysis sample thus comprises 694 participants (aged 10–14 years; Mean [M] = 12.15, Standard Deviation [SD] = 0.84) for whom complete data were available on all core variables included in our network model. This age range represents a critical developmental period spanning late childhood and the onset of early adolescence.

2.2 Procedure and ethical considerations

Data collection was conducted from March to May 2021 under the supervision of professionally trained personnel. The survey, which utilized standardized questionnaires to assess students’ psychological well-being, family dynamics, and peer relationships, was administered in classroom settings. Prior to administration, supervisors thoroughly explained the study’s purpose and procedures to participants. They remained present throughout the survey administration to maintain classroom order, answer queries, and ensure smooth implementation.

The study adhered to strict ethical guidelines aligned with the Helsinki Declaration of 1975 (as revised in 2000). Although direct ethical approval for the current analysis was not required due to the use of existing, de-identified data, the original data collection received approval from the institutional ethics committee (approval number: CAS-WX2021PY-0204). Informed consent was obtained from all participants before their involvement. During data processing, all personal information was anonymized to protect participant privacy. The data preparation phase included recoding reverse-scored items, calculating scores for each research variable, and generating new variables for analysis. A comprehensive review of the entire dataset was conducted to ensure accuracy and data integrity.

2.3 Measures

2.3.1 Demographic information

Participants provided information on their sex, age, academic grade level (Gr), only child status (OCS), and left-behind child status (LB). They also reported on the perceived quality of parental relationships (PR), rated on a 5-point scale from 1 (Very Harmonious) to 5 (Very Disharmonious); desired educational attainment (DEA); and preference for future work location (PFWL).

2.3.2 Generalized anxiety disorder scale

The 7-item Generalized Anxiety Disorder (GAD-7) scale (17) was used to assess anxiety symptoms. The Chinese version uses a 4-point scoring system (0–3), with total scores ranging from 0 to 21 (18). The scale demonstrated good internal consistency in this study (α = .81).

2.3.3 Child behavior checklist

Child problem behaviors were assessed using a 20-item scale derived from the Rule-Breaking Behavior subscale of the Achenbach CBCL (19). Items are rated on a 3-point scale (0–2), with higher scores indicating greater problem behaviors. The Chinese version showed good reliability (α = .80) (20).

2.3.4 Father Involvement questionnaire

The FIQ is a 22-item scale evaluating paternal involvement in areas such as care, academic support, and emotional communication (21). Items are scored from 0 to 4, with higher total scores indicating greater involvement. The scale exhibited excellent reliability in this study (α = .94).

2.3.5 Peer relationship scale

The PRS is an 8-item tool assessing the quality of peer relationships (22). Items are rated on a 5-point scale (scores range 0–32), with higher scores signifying better relationships. It demonstrated good reliability (α = .86) in this study.

2.3.6 Child depression inventory

The CDI is a 27-item scale that evaluates the emotional, cognitive, and behavioral aspects of depression in children (23). Items are scored on a 3-point scale (0–2), with total scores ranging from 0 to 54. The Chinese version of the Children’s Depression Inventory (CDI-C) showed good internal consistency in this survey (α = .88) (24).

2.4 Statistics analysis

Mixed Graphical Models (MGMs) are designed to analyze datasets that include both continuous and discrete variables. They offer a flexible approach suitable for high-dimensional data. In MGMs, a computationally efficient regression-based algorithm is used for model fitting, focusing on the conditional log-likelihood of each variable given the rest. The parameters of MGMs have a natural group structure, and sparsity in the fitted graph is achieved by incorporating a group lasso penalty, which is approximated by a weighted lasso penalty for computational efficiency. Further, we employed a Gaussian Graphical Model (GGM) to delve into CDI structure. We chose a λ-regularization parameter based on the Extended Bayesian Information Criterion (EBIC) for graph fitting. Key centrality measures—strength, closeness, betweenness, and expected influence—were calculated to identify influential nodes. Stability and robustness of our findings were ensured through a case-dropping bootstrap approach. In our study, we used ‘mgm’, ‘bootnet’, and ‘estimateNetworkpackages’ in R.

3 Result

3.1 Descriptive statistics

Table 1 presents descriptive statistics for the key measurements in our study, stratified by demographic variables. For instance, female participants showed slightly higher mean scores on both GAD and CDI compared to male participants. Left-behind children exhibited marginally higher scores on both GAD and CDI compared to non-left-behind children. Notably, parental relationship quality showed a clear trend, with children reporting very harmonious parental relationships having the lowest mean scores on both GAD and CDI, while those reporting not very harmonious relationships had the highest mean scores.

Table 1

Table 1. Descriptive statistics for the measurements on rural children.

3.2 Network analysis using MGMs

A Mixed Graphical Model (MGM) was estimated to explore the relationships between family factors (FIQ, PR), peer relationships (PRS), mental health outcomes (CDI, GAD, CBCL), and demographic variables. The resulting network structure was sparse, with only 13 of 66 possible edges showing non-zero weights after regularization. The network structure is visualized in Figure 1. Several key connections emerged. A strong positive association was found between the Child Depression Inventory (CDI) and the Generalized Anxiety Disorder (GAD) scale, with a weight of 0.55. The Peer Relationship Scale (PRS) was negatively associated with CDI (weight = -0.31) and positively associated with the Father Involvement Questionnaire (FIQ) (weight = 0.22). In turn, FIQ was negatively associated with Parental Relationship quality (PR) (weight = -0.22); it is important to note that the PR scale was reverse-coded, with higher scores indicating less harmonious relationships. Notably, the ‘Left-Behind’ (LB) status variable showed no direct connections to any mental health outcomes in the final network.

Figure 1

Figure 1. Estimated network structure of 694 children based on the mgm package in R. Green/red edges indicate positive/negative weights between nodes. Rings on nodes indicate R² (continuous variable, orange) or categorical accuracy (categorical variable, light blue).

Despite the significant mean differences found in the descriptive t-tests, gender did not emerge as a significant node in the network analysis after the weighted lasso penalty was applied. This suggests that while mean-level differences exist, gender was not a robust structural connector to other variables once the full system of interactions was modeled.

Centrality indices—strength, closeness, betweenness, and expected influence—were calculated to identify the most influential nodes in the network (see Figure 2). The CDI node exhibited the highest centrality in terms of strength, closeness, and betweenness, indicating its critical role in network connectivity. GAD also demonstrated substantial influence, followed by PRS. CDI did not, however, have the highest expected influence, which accounts for the direction and weight of connections. The stability of the network was assessed using a case-dropping bootstrap approach (see Figure 3). The Correlation Stability (CS) coefficient indicated high stability for strength (CS[cor = 0.7] = 0.75) and expected influence (CS[cor = 0.7] = 0.62), both of which are above the recommended 0.5 threshold. The betweenness index, however, was found to be unstable (CS[cor = 0.7] = 0), likely due to the sparse nature of the network and multiple disconnections between variables.

Figure 2

Figure 2. Centrality indices for the nodes of the present network, including those for strength, betweenness, closeness, and expected influence. The full names of the abbreviations can be found in Figure 1.

Figure 3

Figure 3. The average correlation coefficients between selected cases and the full sample for the centrality indices of networks. Lines indicate the means and areas ranging from the 2.5th quantile to the 97.5th quantile.

3.3 Depression symptom network analysis

Given the high centrality of the CDI node, a secondary Gaussian Graphical Model (GGM) was estimated to explore the internal structure of depression symptoms, along with their specific connections to Peer Relationships (PRS) and Parental Relationships (PR). This network is presented in Figure 4. The resulting network was fully connected, with all 21 edges bearing significant weights, indicating no regularization was necessary. A notable finding within this symptom-level network was a strong negative relationship between peer relationships (PRS) and anhedonia (weight = -0.27).

Figure 4

Figure 4. Estimated network structure of 694 children based on the estimateNetwork package in R. Blue/red edges indicate positive/negative weights between nodes.

Centrality analysis of the depression symptom network (Figure 5) revealed that Negative Mood (NM) was the most central node, exhibiting the highest scores across all calculated metrics (strength, closeness, betweenness, and eigenvector centrality). This underscores its pivotal role within the depression symptom structure. Furthermore, when comparing the two social variables, the peer relationship (PRS) node was more central than the parental relationship (PR) node according to strength, closeness, and betweenness metrics.

Figure 5

Figure 5. Centrality indices for the nodes of the present network, including those for strength, betweenness, closeness, and expected influence.

The stability of this second network (Figure 6) was found to be highly robust. The CS coefficients were high for betweenness (CS[cor = 0.7] = 0.672) and exceptionally high for both strength and eigenvector centrality (CS[cor = 0.7] = 0.75), indicating that the network attributes are reliable.

Figure 6

Figure 6. The average correlation coefficients between selected cases and the full sample for the centrality indices of networks. Lines indicate the means and areas ranging from the 2.5th quantile to the 97.5th quantile.

4 Discussion

Our network analysis of 694 rural Chinese children offers a nuanced perspective on the architecture of mental health during early adolescence, highlighting the central role of peer relationships. Grounded in social development theory, our central finding is that peer relationships exerted a stronger relationship to psychological well-being than family factors. This aligns with an ecological systems perspective (25), suggesting that in the specific context of rural China—where parental migration is common—the peer microsystem may become a primary source of social and emotional support. In this integrated discussion, we explore how this central finding contextualizes our other key results, including the strong comorbidity between depression and anxiety, the centrality of negative mood, and the notable absence of a direct link between left-behind status and mental health outcomes.

The preeminence of peer relationships over family factors in our network is a critical finding. This may be amplified by the unique social structures of rural Chinese schools, which often feature more stable and cohesive populations compared to more transient urban environments (26). These long-term, stable peer cohorts can foster deep connections that serve as a primary and reliable source of emotional support, becoming particularly salient when traditional family support is less consistent due to parental migration. Furthermore, this finding may reflect an ongoing sociocultural transformation in China. While traditional Chinese culture emphasizes familial hierarchy (27), rapid modernization may be elevating the role of peer groups as a powerful force for socialization and identity formation, even in rural settings.

One of the most intriguing findings of our study is the minimal direct impact of the left-behind experience on mental health outcomes. While extensive literature documents the adverse psychological effects of parental migration (28, 29), our results suggest a more complex reality. Rather than attributing this to an unmeasured construct like resilience, our network analysis points to a potential explanation within the data itself: the powerful, buffering influence of peer relationships. It is highly plausible that in these rural communities, supportive and stable peer networks serve as a critical protective factor, mitigating the potential loneliness or distress associated with parental absence. Our finding that peer relationships are highly central to well-being supports this interpretation. This suggests the relation to being “left-behind” is not absent, but is likely indirect and mediated by the quality of these more proximal peer and remaining familial connections, a hypothesis that future longitudinal and mediation analyses should explore.

Beyond the social context, our findings illuminate the internal structure of psychopathology in this population. The network revealed a robust positive edge between depression and anxiety, corroborating the high comorbidity consistently found in adolescent populations (30, 31). This strong link underscores that these are not isolated issues but rather intertwined conditions that likely share underlying vulnerabilities and reinforce one another. Furthermore, our more granular analysis of the depression network itself identified “Negative Mood” as its most central node. This suggests that negative affect is a core component of the depressive experience for these children, not just a peripheral symptom. This finding has significant clinical implications, suggesting that interventions targeting this core affective component may be more effective at destabilizing the entire depressive network than interventions focused on other, less central symptoms.

Our findings offer clear, actionable implications for mental health interventions in rural China. First, the primacy of peer relationships suggests that support systems should expand beyond traditional family-centric models. School-based interventions that focus on fostering positive peer interactions, improving social skills, and establishing peer-support programs may be particularly effective and resource-efficient (32). Second, the centrality of Negative Mood underscores the need for interventions that directly target core depressive symptoms and emotional well-being, rather than focusing solely on behavioral issues (33). Psychoeducational programs that teach children to identify and manage negative affect could provide a critical foundation for improved mental health (34).

While this study provides valuable insights, its findings should be interpreted in light of several limitations, which also point to important directions for future research. First, the cross-sectional design precludes inferences of causality (35); longitudinal studies are needed to untangle the directional relationships between peer quality and mental health. Second, our sample was restricted to fifth- and sixth-grade students from specific provinces. While diverse, these regions predominantly represent rural areas with specific developmental profiles (mostly medium to low development, with the exception of Guangdong), and the findings may not generalize to other age groups, highly urbanized regions, or areas with different socioeconomic characteristics (36). Third, measurement limitations include the use of a single-item measure for parental relationships and the lack of data on the duration of the left-behind experience or the specific role of maternal involvement. Furthermore, as with all self-report measures, the data may be subject to social desirability bias or recall bias, particularly regarding sensitive topics like family dynamics and mental health symptoms. Future research should utilize more comprehensive measures and collect primary data on the temporal dimensions of parental absence. Finally, while network analysis is robust for exploring systemic relationships, future studies could use Structural Equation Modeling (SEM) to formally test the mediation hypotheses proposed here (e.g., peer relationships mediating the effect of left-behind status).

5 Conclusion

This study used network analysis to reveal that the mental health of rural Chinese early adolescents is a complex system wherein peer relationships are a more central role than family factors. Depression and anxiety were strongly comorbid, and “Negative Mood” was identified as a core component of the depression network. The impact of left-behind status appears to be indirect, likely buffered by the strength of these proximal peer dynamics. These findings highlight a critical need for interventions in rural communities to shift toward strengthening positive peer interactions and directly addressing core affective symptoms.

Data availability statement

Publicly available datasets were analyzed in this study. This data can be found here: http://doi.org/10.57760/sciencedb.j00001.00464.

Ethics statement

The studies involving humans were approved by (approval number: CAS-WX2021PY-0204). The studies were conducted in accordance with the local legislation and institutional requirements. Written informed consent for participation in this study was provided by the participants’ legal guardians/next of kin.

Author contributions

ZH: Data curation, Writing – original draft, Methodology. HL: Supervision, Writing – review & editing, Project administration.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the Science and Technology Plan Project of Wenzhou (Grant No. Y20240628).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was used in the creation of this manuscript. Generative AI was used: Primarily for language editing and polishing of the manuscript. Its application was limited to improving grammar, clarity, and conciseness. The authors reviewed, edited, and approved all AI-assisted text to ensure it accurately reflects their original work and intent. The AI was not used for data analysis, interpretation of results, or generating scientific conclusions.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Abbreviations

CBCL, Child Behavior Checklist; CDI, Child Depression Inventory; CFI, Comparative Fit Index; CS, Consistency Selection; DEA, Desired Educational Attainment; EBIC, Extended Bayesian Information Criterion; FIQ, Father Involvement Questionnaire; GAD, Generalized Anxiety Disorder Scale; GAD-7, Generalized Anxiety Disorder 7-item scale; GGM, Gaussian Graphical Model; Gr, Academic Grade Level; LB, Left-Behind child status; MGMs, Mixed Graphical Models; NM, Negative Mood; OCS, Only Child Status; PFWL, Preference for Future Work Location; PR, Quality of Parental Relationships; PRS, Peer Relationship Scale; RMSEA, Root Mean Square Error of Approximation; SD, Standard Deviation; TLI, Tucker-Lewis Index.

References

1. Jaworska N and MacQueen G. Adolescence as a unique developmental period. J Psychiatry Neurosci. (2015) 40:291–3. doi: 10.1503/jpn.150268

PubMed Abstract | Crossref Full Text | Google Scholar

2. Ruan QN, Shen GH, Xu S, Xu D, and Yan WJ. Depressive symptoms among rural left-behind children and adolescents in China: a large-scale cross-sectional study. BMC Public Health. (2024) 24:3160. doi: 10.1186/s12889-024-20699-z

PubMed Abstract | Crossref Full Text | Google Scholar

3. Morales DA, Barksdale CL, and Beckel-Mitchener AC. A call to action to address rural mental health disparities. J Clin Transl Sci. (2020) 4:463–7. doi: 10.1017/cts.2020.42

PubMed Abstract | Crossref Full Text | Google Scholar

4. Wikle JS and Hoagland A. Adolescent interactions with family and emotions during interactions: Variation by family structure. J Fam Psychol. (2020) 34:544. doi: 10.1037/fam0000625

PubMed Abstract | Crossref Full Text | Google Scholar

5. Zhao Q, Yu X, Wang X, and Glauben T. The impact of parental migration on children’s school performance in rural China. China Econ Rev. (2014) 31:43–54. doi: 10.1016/j.chieco.2014.07.013

Crossref Full Text | Google Scholar

6. Wang Y, Yang S, Wang F, and Liu Z. Long-term effects of left-behind experience on adult depression: Social trust as mediating factor. Front Public Health. (2022) 10:957324. doi: 10.3389/fpubh.2022.957324

PubMed Abstract | Crossref Full Text | Google Scholar

7. Hadfield K, Al-Soleiti M, Dajani R, Mareschal I, and Panter-Brick C. Father involvement and child development: A prospective study of Syrian refugee families. J Child Fam Stud. (2024) 33:1029–42. doi: 10.1007/s10826-024-02809-y

Crossref Full Text | Google Scholar

8. Giletta M, Choukas-Bradley S, Maes M, Linthicum KP, Card NA, and Prinstein MJ. A meta-analysis of longitudinal peer influence effects in childhood and adolescence. Psychol Bull. (2021) 147:719. doi: 10.1037/bul0000329

PubMed Abstract | Crossref Full Text | Google Scholar

9. Delgado E, Serna C, Martínez I, and Cruise E. Parental attachment and peer relationships in adolescence: A systematic review. Int J Environ Res Public Health. (2022) 19:1064. doi: 10.3390/ijerph19031064

PubMed Abstract | Crossref Full Text | Google Scholar

10. Ruan QN, Liu L, Shen GH, Wu YW, and Yan WJ. Alexithymia and peer victimisation: interconnected pathways to adolescent non-suicidal self-injury. BJPsych Open. (2024) 10:e46. doi: 10.1192/bjo.2023.653

PubMed Abstract | Crossref Full Text | Google Scholar

11. Butler N, Quigg Z, Bates R, Jones L, Ashworth E, Gowland S, et al. The contributing role of family, school, and peer supportive relationships in protecting the mental wellbeing of children and adolescents. School Ment Health. (2022) 14:776–88. doi: 10.1007/s12310-022-09502-9

PubMed Abstract | Crossref Full Text | Google Scholar

12. Ruan QN, Chen YH, and Yan WJ. A network analysis of difficulties in emotion regulation, anxiety, and depression for adolescents in clinical settings. Child Adolesc Psychiatry Ment Health. (2023) 17:29. doi: 10.1186/s13034-023-00574-2

PubMed Abstract | Crossref Full Text | Google Scholar

13. Sánchez-Hernández MO, Holgado-Tello FP, and Carrasco MÁ. Network analysis of internalizing and externalizing symptoms in children and adolescents. Psicothema. (2023) 35:66. doi: 10.7334/psicothema2022.17

PubMed Abstract | Crossref Full Text | Google Scholar

14. Long E, Gardani M, McCann M, Sweeting H, Tranmer M, and Moore L. Mental health disorders and adolescent peer relationships. Soc Sci Med. (2020) 253:112973. doi: 10.1016/j.socscimed.2020.112973

PubMed Abstract | Crossref Full Text | Google Scholar

15. Brown BB and Larson J. Peer relationships. Adolesc Handb Adolesc Psychol. (2009) 2:74–103.

Google Scholar

16. Yuan F, Wang Y, Li C, Chen Z, Li C, Li C, et al. The database on rural children’s mental health and its influencing factors. Sci Data Bank. (2023).

Google Scholar

17. Spitzer RL, Kroenke K, Williams JB, and Löwe B. A brief measure for assessing generalized anxiety disorder: the GAD-7. Arch Intern Med. (2006) 166:1092–7. doi: 10.1001/archinte.166.10.1092

PubMed Abstract | Crossref Full Text | Google Scholar

18. Zeng QZ, He YL, Liu H, Miao JM, Chen JX, Xu HN, et al. Reliability and validity of Chinese version of the Generalized Anxiety Disorder 7-item (GAD-7) scale in screening anxiety disorders in outpatients from traditional Chinese internal department. Chin Ment Health J. (2013). Available online at: https://psycnet.apa.org/record/2013-12337-001 (Accessed November 19, 2025).

Google Scholar

19. Achenbach TM and Rescorla LA. Child Behavior Checklist for Ages 6-18 (2001). VT: University of Vermont Burlington. Available online at: https://reachoutandread.org/wp-content/uploads/2023/06/CBCL-1-Summary.pdf (Accessed November 19, 2025).

Google Scholar

20. Leung PWL, Kwong SL, Tang CP, Ho TP, Hung SF, Lee CC, et al. Test–retest reliability and criterion validity of the Chinese version of CBCL, TRF, and YSR. J Child Psychol Psychiatry. (2006) 47:970–3. doi: 10.1111/j.1469-7610.2005.01570.x

PubMed Abstract | Crossref Full Text | Google Scholar

21. Mautone JA, Marcelle E, Tresco KE, and Power TJ. ASSESSING THE QUALITY OF PARENT–TEACHER RELATIONSHIPS FOR STUDENTS WITH ADHD. Psychol Sch. (2015) 52:196–207. doi: 10.1002/pits.21817

PubMed Abstract | Crossref Full Text | Google Scholar

22. DeWalt DA, Thissen D, Stucky BD, Langer MM, DeWitt EM, Irwin DE, et al. PROMIS Pediatric Peer Relationships Scale: development of a peer relationships item bank as part of social health measurement. Health Psychol. (2013) 32:1093. doi: 10.1037/a0032670

PubMed Abstract | Crossref Full Text | Google Scholar

23. Wierzbicki M. A parent form of the Children’s Depression Inventory: Reliability and validity in nonclinical populations. J Clin Psychol. (1987) 43:390–7. doi: 10.1002/1097-4679(198707)43:4%3C390::AID-JCLP2270430409%3E3.0.CO;2-S

PubMed Abstract | Crossref Full Text | Google Scholar

24. Wu WF, Lu YB, Tan FR, and Yao SQ. Reliability and validity of the Chinese version of Children’s Depression Inventory. Chin Ment Health J. (2010) 24:775–9. Available online at: https://psycnet.apa.org/record/2010-22384-014 (Accessed November 19, 2025).

Google Scholar

25. Bronfenbrenner U. Ecological models of human development. Int Encycl Educ. (1994) 3:37–43.

Google Scholar

26. Zhang D. The rural-urban divide, intergroup relations, and social identity formation of rural migrant children in a Chinese urban school. Int Stud Sociol Educ. (2018) 27:60–77. doi: 10.1080/09620214.2017.1394204

Crossref Full Text | Google Scholar

27. Han Q. The ties that bind: An overview of traditional Chinese family ethics. J Chin Stud. (2012) 1:85–99.

Google Scholar

28. Zhao C, Wang F, Li L, Zhou X, and Hesketh T. Long-term impacts of parental migration on Chinese children’s psychosocial well-being: mitigating and exacerbating factors. Soc Psychiatry Psychiatr Epidemiol. (2017) 52:669–77. doi: 10.1007/s00127-017-1386-9

PubMed Abstract | Crossref Full Text | Google Scholar

29. Raturi R and Cebotari V. The impact of parental migration on psychological well-being of children in Ghana. J Ethn Migr Stud. (2023) 49:192–211. doi: 10.1080/1369183X.2022.2076070

Crossref Full Text | Google Scholar

30. Konac D, Young KS, Lau J, and Barker ED. Comorbidity between depression and anxiety in adolescents: bridge symptoms and relevance of risk and protective factors. J Psychopathol Behav Assess. (2021) 43:583–96. doi: 10.1007/s10862-021-09880-5

PubMed Abstract | Crossref Full Text | Google Scholar

31. Cummings CM, Caporino NE, and Kendall PC. Comorbidity of anxiety and depression in children and adolescents: 20 years after. Psychol Bull. (2014) 140:816. doi: 10.1037/a0034733

PubMed Abstract | Crossref Full Text | Google Scholar

32. Fortuna KL, Solomon P, and Rivera J. An update of peer support/peer provided services underlying processes, benefits, and critical ingredients. Psychiatr Q. (2022) 93:571–86. doi: 10.1007/s11126-022-09971-w

PubMed Abstract | Crossref Full Text | Google Scholar

33. Weijers A, Rasing S, Creemers D, Vermulst A, Schellekens AFA, and Westerhof GJ. The relationship between depressive symptoms, general psychopathology, and well-being in patients with major depressive disorder. J Clin Psychol. (2021) 77:1472–86. doi: 10.1002/jclp.23083

PubMed Abstract | Crossref Full Text | Google Scholar

34. Luo H. Emotional management and mental health promotion in young children: an empirical study of educational intervention. Eurasian J Educ Res EJER. (2023) 2023:212. Available online at: https://search.ebscohost.com/login.aspx?direct=true&profile=ehost&scope=site&authtype=crawler&jrnl=1302597X&AN=176490414&h=ZSAJZwcnMuBzn5ZhXeHymoyiNyLMyuefvRZ%2BaXP8HR7jKr9xxXWPRbMBjBAYGCAI92Png4gcu1hfFN%2FdJp11Hw%3D%3D&crl=c (Accessed November 19, 2025).

Google Scholar

35. Savitz DA and Wellenius GA. Can cross-sectional studies contribute to causal inference? It depends. Am J Epidemiol. (2023) 192:514–6. doi: 10.1093/aje/kwac037

PubMed Abstract | Crossref Full Text | Google Scholar

36. Chen Z, Hu B, Liu X, Becker B, Eickhoff SB, Miao K, et al. Sampling inequalities affect generalization of neuroimaging-based diagnostic classifiers in psychiatry. BMC Med. (2023) 21:241. doi: 10.1186/s12916-023-02941-4

PubMed Abstract | Crossref Full Text | Google Scholar