Association between parenting styles and child and adolescent mental health: The mediation effect of family functioning and self-concept

Abstract

BACKGROUND

Parenting styles are closely associated with adolescent mental health; however, the mediating roles of self-concept and family functioning in this relationship remain insufficiently understood.

AIM

To examine the associations of parental rearing styles with mental health symptoms in Chinese children and adolescents and to determine whether self-concept and family functioning transmit the indirect effects.

METHODS

This observational study included 4433 children and adolescents aged 9-18 years. Parenting styles, mental health symptoms, self-concept, and family functioning were assessed using the Egna Minnen Betraffande Uppfostran questionnaire, the Mental Health Test, the Piers-Harris Children’s Self-Concept Scale, and the McMaster Family Assessment Device, respectively. Multivariable logistic regression analyses were used to evaluate the associations between parenting styles and mental health symptoms, and mediation analyses were conducted to examine the indirect effects of self-concept and family functioning.

RESULTS

The overall prevalence of mental health symptoms was 6.5%. Each 1-SD increase in paternal and maternal emotional warmth was significantly associated with lower odds of mental health symptoms [paternal: Odds ratio (OR) = 0.52; maternal: OR = 0.58; both P < 0.001]. In contrast, maladaptive parenting dimensions, including punishment, rejection, and overprotection, were significantly associated with increased odds of mental health symptoms (adjusted OR range per 1-SD, 1.75-2.06; all P < 0.001). Both self-concept and family functioning significantly mediated these associations. The mediation proportions for paternal and maternal emotional warmth through self-concept were 121.48% and 121.15%, respectively, indicating suppression, whereas the corresponding mediation proportions through family functioning were 42.78% and 55.8%, respectively. Significant indirect effects of maladaptive parenting styles on mental health symptoms were also identified through both mediating pathways (all P < 0.001).

CONCLUSION

Parenting styles showed significant associations with mental health symptoms among children and adolescents, with self-concept and family functioning serving as important mediating mechanisms. Longitudinal designs are required before causal inferences can be made.

Key Words: Parenting styles; Mental health symptoms; Self-concept; Family functioning; Children and adolescents

Core Tip: In this cross-sectional study of 4433 Chinese adolescents, higher levels of emotional warmth from either parent were associated with lower odds of mental health symptoms, whereas harsh parenting predicted a greater symptom burden. Self-concept mediated over 121% of the protective effect of emotional warmth (a suppression pattern), and family functioning explained approximately 40%. Given that both self-concept and family functioning are modifiable, these factors may constitute promising intervention targets to improve adolescent mental health.

INTRODUCTION

The prevalence of mental health problems among children and adolescents has risen substantially in recent decades and has become a major global public health concern. Estimates from the World Health Organization indicate that approximately 10%-20% of children and adolescents worldwide experience diagnosable mental health disorders[1]. In China, about 17.6% of young people reportedly exhibit emotional or behavioral problems[2]. As adolescence represents a critical developmental stage characterized by rapid psychological, emotional, and social maturation, identifying modifiable determinants of mental health has become a key priority for prevention and early intervention.

Among the environmental influences on adolescent development, the family environment constitutes one of the most salient microsocial systems shaping psychological well-being. Parenting styles, typically defined along dimensions such as emotional warmth, rejection, punishment, and overprotection, have been consistently associated with adolescent mental health outcomes[3]. Substantial evidence indicates that supportive parenting fosters self-esteem, emotional security, and adaptive psychosocial functioning[4]. In contrast, maladaptive parenting behaviors, including punitive discipline and parental rejection, have been linked to increased risks of both internalizing and externalizing psychopathology[5-7]. Beyond discrete parenting behaviors, family functioning has increasingly been recognized as a key contextual mechanism through which parenting styles may affect adolescent mental health[8]. Dysfunctional family functioning (characterized by poor communication, emotional disengagement, and ineffective problem-solving) has been associated with elevated risks of depression and other psychological difficulties in adolescents[9].

In addition to family contextual factors, self-concept appears to be an important psychological mechanism underlying the association between parenting styles and mental health outcomes. Self-concept refers to an individual’s cognitive and evaluative perception of the self and plays a central role in emotion regulation and psychological adjustment. According to Roger’s humanistic theory[10], unconditional positive regard from caregivers facilitates the development of a coherent and integrated self-concept, thereby enhancing psychological resilience and overall psychological well-being. In contrast, conditional acceptance and parental criticism may distort self-perception, leading to self-incongruence and heightened vulnerability to psychological distress.

Despite extensive research on parenting styles, family functioning, self-concept, and adolescent mental health, several critical gaps remain. First, the concurrent mediating roles of self-concept and family functioning have seldom been tested within an integrated analytical framework, limiting understanding of the complex pathways linking parenting styles to adolescent mental health. Second, potential differences between paternal and maternal parenting styles in their direct and indirect effects on mental health outcomes remain insufficiently elucidated. Addressing these gaps may yield important insights into the psychological and familial pathways underlying adolescent mental health problems. Accordingly, the present cross-sectional study examined the associations between parenting styles and mental health symptoms in a large sample of Chinese children and adolescents and tested the mediating roles of self-concept and family functioning to clarify the mechanisms underlying these associations.

MATERIALS AND METHODS

Study design and participants

This observational study was conducted from September 2025 to December 2025 in five Chinese provinces: Shaanxi Province, Guangdong Province, Hubei Province, Hebei Province, and Shanxi Province. Among the 4433 included participants, 3079 (69.5%) were from Shaanxi Province, 538 (12.1%) from Guangdong Province, 438 (9.9%) from Hubei Province, 238 (5.4%) from Hebei Province, and 140 (3.2%) from Shanxi Province. A collaborative research network was established by recruiting regional investigators online across the five provinces. These investigators used purposive convenience sampling to identify partner primary and secondary schools within their jurisdictions, based on school willingness to participate and feasibility of site access. Within each participating school, students in grades 4-12 were invited to complete a centralized electronic questionnaire under supervision in school computer laboratories.

Participants were excluded if they refused informed consent after receiving a detailed explanation of study procedures or could not understand the questionnaire and complete the online survey independently (e.g., due to severe cognitive impairment). A total of 7575 questionnaires were initially collected. Subsequent exclusions were applied for missing data on any study variable, implausible age values (< 8 years or > 18 years), or validity scale scores ≥ 7 on the Mental Health Test (MHT), indicative of potentially unreliable responses. The participant screening and enrollment process is shown in Figure 1.

Figure 1 Flow diagram of the screening and enrollment of study participants. MHT: Mental Health Test.

The study protocol was approved by the Medical Ethics Committee of Shaanxi Provincial People’s Hospital (Approval No. 2025R039). All procedures complied with the ethical principles of the Declaration of Helsinki (1975, revised in 2000). Written informed consent was obtained from all participants and their legal guardians before enrollment.

Measures

Mental health symptoms: Mental health symptoms were assessed using the MHT, a widely used 100-item self-report instrument developed by Zhou[11]. The MHT assesses eight dimensions of psychological symptoms: Learning anxiety, social anxiety, loneliness tendency, self-blame tendency, allergic tendency, physical symptoms, phobic tendency, and impulsive tendency, as well as a 10-item validity scale designed to detect response inconsistency. Each item is scored dichotomously (“yes” = 1; “no” = 0), with higher total scores indicating poorer mental health. In accordance with Chinese normative standards, total MHT scores ≥ 65 were classified as indicative of mental health symptoms.

Parental rearing styles: Parenting styles were measured using the Chinese version of the Egna Minnen Betraffande Uppfostran (EMBU) questionnaire, originally developed by Perris et al[12] and later revised by Dong-Mei Yue for use in Chinese populations. The paternal version comprises 58 items across six dimensions: Emotional warmth, punishment and severity, overinvolvement, parental preference, rejection and denial, and overprotection. The maternal version contains 57 items and covers five dimensions, excluding overinvolvement. All items are rated on a 4-point Likert scale from 1 (“never”) to 4 (“always”), with higher subscale scores indicating stronger endorsement of the corresponding parenting behavior.

Family functioning: Family functioning was assessed using the McMaster Family Assessment Device (FAD), a 60-item self-report instrument developed by Epstein et al[13] and later adapted into Chinese by Pei-Yi Liu. The FAD evaluates six core dimensions of family functioning (problem solving, communication, roles, affective responsiveness, affective involvement, and behavioral control) together with a general functioning dimension. Each item was rated on a 4-point Likert scale from 1 (“strongly agree”) to 4 (“strongly disagree”), with higher total scores reflecting poorer family functioning.

Children’s self-concept: Children’s self-concept was measured using the Piers-Harris Children’s Self-Concept Scale (PHCSS), an 80-item dichotomous self-report measure requiring “yes” or “no” responses[14]. The PHCSS yields scores across six domains: Behavior, intellectual and school status, physical appearance and attributes, anxiety, popularity, and happiness and satisfaction. Chinese urban norms were established by Lin-Yan Su and colleagues.

Covariates: Sociodemographic characteristics were obtained using a structured self-administered questionnaire and included age, sex, grade, residential area (urban vs rural/county), only-child status, left-behind status (both parents present, mother absent, father absent, or both absent), parental marital status (married vs divorced), parental educational attainment (junior high school or below, high school/secondary vocational school, associate degree, bachelor’s degree, or master’s degree and above), and household income level (affluent, moderate, below average, or financially disadvantaged). These variables were included as potential confounders based on prior research on determinants of child and adolescent mental health.

Statistical analysis

Baseline characteristics were compared between participants with and without mental health symptoms. Normally distributed continuous variables are presented as means ± SD, and categorical variables as n (%). Group differences were examined using independent-samples t tests for continuous variables and χ² tests for categorical variables.

To examine the independent associations of parenting styles, self-concept, and family functioning with mental health symptoms, multivariable logistic regression models were fitted, and odds ratios (ORs) with 95% confidence intervals (CIs) were estimated. For the EMBU parenting dimensions, each subscale was standardized to a z-score (0 ± 1) before inclusion in the models. All models adjusted for age, sex, only-child status, residential area, province, parental marital status, paternal educational attainment, maternal educational attainment, household income, and left-behind status.

Mediation analyses were conducted using a regression-based path decomposition framework with bootstrap inference[14]. Indirect effects (a × b) and direct effects (c′) were estimated from 5000 bootstrap resamples with 95% bias-corrected CIs; mediation was deemed present when the CI for a × b did not include zero. Both unadjusted and adjusted models were estimated. The proportion mediated was calculated as (a × b)/c; values exceeding 100% were interpreted as suppression, indicating opposite directions of indirect and direct effects[15].

All statistical analyses were carried out using R Statistical Software (version 4.2.2; R Foundation for Statistical Computing, Vienna, Austria) and the Free Statistics platform (version 2.3; Beijing Free Clinical Medical Technology Co., Ltd., Beijing, China). All tests were two-sided, and P < 0.05 were considered statistically significant.

RESULTS

Characteristics of the participants

A total of 4433 children and adolescents were included in the final analysis, with a mean age of 13.2 years; 2168 (48.9%) were female. The overall prevalence of mental health symptoms was 6.5% (n = 290). Detailed baseline characteristics by mental health symptom status are presented in Supplementary Table 1. Demographic characteristics were comparable between included and excluded participants (all P > 0.05; Supplementary Table 2).

Relative to participants without mental health symptoms, those with symptoms were older (14.0 years vs 13.1 years), more likely to be female (60.3% vs 48.1%), and had higher rates of parental divorce (14.1% vs 8.0%) and financial difficulties (12.4% vs 5.7%). No significant between-group differences were observed for only-child status, left-behind status, or parental educational attainment (all P > 0.05).

Multivariable regression analysis

Associations of parenting styles, self-concept, and family functioning with mental health symptoms are summarized in Tables 1, 2 and 3.

Table 1 Association of parental rearing style with mental health symptoms among participants.

Variable	Total	Event	Crude model		Adjusted model1
			OR (95%CI)	P value	OR (95%CI)	P value
Paternal emotional warmth	4433	290	0.94 (0.93-0.95)	< 0.001	0.94 (0.93-0.96)	< 0.001
Paternal punishment and severity	4433	290	1.11 (1.1-1.13)	< 0.001	1.12 (1.1-1.13)	< 0.001
Paternal over involvement	4433	290	1.16 (1.13-1.19)	< 0.001	1.15 (1.13-1.18)	< 0.001
Paternal preference for the child	4433	290	0.99 (0.96-1.02)	0.408	0.97 (0.93-1.01)	0.18
Paternal rejection and denial	4433	290	1.25 (1.21-1.29)	< 0.001	1.24 (1.2-1.28)	< 0.001
Paternal over protection	4433	290	1.31 (1.26-1.37)	< 0.001	1.31 (1.26-1.37)	< 0.001
Maternal emotional warmth	4433	290	0.95 (0.94-0.96)	< 0.001	0.95 (0.94-0.96)	< 0.001
Maternal over interference and over protection	4433	290	1.11 (1.09-1.12)	< 0.001	1.1 (1.09-1.12)	< 0.001
Maternal rejection and denial	4433	290	1.18 (1.15-1.2)	< 0.001	1.17 (1.14-1.19)	< 0.001
Maternal punishment and severity	4433	290	1.15 (1.12-1.17)	< 0.001	1.15 (1.12-1.17)	< 0.001
Maternal preference for the child	4433	290	0.99 (0.96-1.02)	0.453	0.97 (0.93-1.02)	0.219

¹Adjusted for age, gender, residence, only child, left behind status, parental marital status, father’s education, mother’s education, family income.

OR: Odds ratio; CI: Confidence interval.

Table 2 Association of children’s self-concept with mental health symptoms among participants - Piers-Harris Children’s Self-Concept Scale.

Variable	Total	Event	Crude model		Adjusted model1
			OR (95%CI)	P value	OR (95%CI)	P value
Behavior	4433	290	0.68 (0.66-0.71)	< 0.001	0.68 (0.66-0.71)	< 0.001
Intellectual and school status	4433	290	0.72 (0.69-0.75)	< 0.001	0.72 (0.69-0.76)	< 0.001
Physical appearance and attributes	4433	290	0.71 (0.67-0.74)	< 0.001	0.72 (0.68-0.75)	< 0.001
Anxiety	4433	290	0.5 (0.47-0.54)	< 0.001	0.49 (0.45-0.52)	< 0.001
Popularity	4433	290	0.51 (0.48-0.55)	< 0.001	0.51 (0.48-0.55)	< 0.001
Happiness and satisfaction	4433	290	0.50 (0.45-0.54)	< 0.001	0.52 (0.47-0.56)	< 0.001
PHCSS total score	4433	290	0.86 (0.85-0.87)	< 0.001	0.86 (0.85-0.87)	< 0.001

¹Adjusted for age, gender, residence, only child, left behind status, parental marital status, father’s education, mother’s education, family income.

OR: Odds ratio; CI: Confidence interval; PHCSS, Piers-Harris Children's Self-Concept Scale.

Table 3 Association of family functioning with mental health symptoms among participants - family assessment device.

Variable	Total	Event	Crude model		Adjusted model1
			OR (95%CI)	P value	OR (95%CI)	P value
Problem solving	4433	290	1.04 (1.01-1.08)	0.006	1.05 (1.01-1.08)	0.006
Communication	4433	290	1.23 (1.18-1.27)	< 0.001	1.21 (1.17-1.26)	< 0.001
Roles	4433	290	1.16 (1.13-1.19)	< 0.001	1.15 (1.12-1.19)	< 0.001
Affective responsiveness	4433	290	1.29 (1.23-1.35)	< 0.001	1.26 (1.20-1.32)	< 0.001
Affective involvement	4433	290	1.14 (1.11-1.17)	< 0.001	1.15 (1.12-1.18)	< 0.001
Behavioral control	4433	290	1.15 (1.11-1.20)	< 0.001	1.15 (1.10-1.19)	< 0.001
General functioning	4433	290	1.15 (1.12-1.18)	< 0.001	1.15 (1.12-1.18)	< 0.001

¹Adjusted for age, gender, residence, only child, left behind status, parental marital status, father’s education, mother’s education, family income.

OR: Odds ratio; CI: Confidence interval.

After adjustment for potential confounders, parental emotional warmth consistently emerged as a protective factor. Each 1-SD increase in paternal emotional warmth (OR = 0.52; 95%CI: 0.45-0.59) and maternal emotional warmth (OR = 0.58; 95%CI: 0.52-0.66) was associated with significantly lower odds of mental health symptoms (both P < 0.001).

In contrast, several maladaptive parenting dimensions were associated with increased odds of mental health symptoms. Each 1-SD increase in paternal punishment and severity (OR = 1.82; 95%CI: 1.66-1.99), rejection and denial (OR = 1.93; 95%CI: 1.75-2.13), and overprotection (OR = 2.01; 95%CI: 1.80-2.25), as well as maternal punishment and severity (OR = 1.75; 95%CI: 1.60-1.92), rejection and denial (OR = 1.97; 95%CI: 1.78-2.17), and overinterference and overprotection (OR = 2.06; 95%CI: 1.85-2.30), was positively associated with mental health symptoms (all P < 0.001). Parental preference for the child was not significantly associated with mental health symptoms (Supplementary Tables 3 and 4).

Self-concept showed robust inverse associations with mental health symptoms across all domains. Anxiety demonstrated the strongest association (OR = 0.49; 95%CI: 0.45-0.52), indicating that more positive emotional self-perceptions were strongly related to lower psychological risk. Significant inverse associations were also observed for happiness and satisfaction, popularity, behavior, intellectual and school status, and physical appearance and attributes. Higher total PHCSS scores were consistently associated with reduced odds of mental health symptoms (OR = 0.86; 95%CI: 0.85-0.87; P < 0.001).

Poorer family functioning was consistently associated with a higher likelihood of mental health symptoms. Among the family functioning dimensions, affective responsiveness showed the strongest association (OR = 1.26; 95%CI: 1.20-1.32), followed by communication (OR = 1.21; 95%CI: 1.17-1.26), whereas problem solving showed a weaker but still significant association (OR = 1.05; 95%CI: 1.01-1.08; P = 0.006). These findings indicate that both individual psychological resources and family relational contexts are closely linked to mental health outcomes in children and adolescents.

Mediation analysis

Mediation analyses were conducted to further elucidate the psychological and familial pathways linking parenting styles with mental health symptoms (Figure 2; complete numerical results are provided in Supplementary Tables 5 and 6).

Figure 2 Mediation effects of self-concept and family functioning on the associations between parental rearing styles and mental health symptoms (adjusted model). A: Indirect effects with 95% bootstrap confidence intervals; B: Mediation proportions with 95% bootstrap confidence intervals. Blue squares represent paternal subscales; red circles represent maternal subscales. All estimates are adjusted for age, gender, province, residence, only child, left behind status, parental marital status, father’s education, mother’s education, family income. P: Parental; M: Maternal; CI: Confidence interval.

Self-concept showed substantial mediating effects in these associations. Higher levels of paternal and maternal emotional warmth were indirectly associated with lower levels of mental health symptoms via enhanced self-concept, with indirect effects of -0.0131 and -0.0107 and mediation proportions of 121.48% and 121.15%, respectively (both P < 0.001). These results suggest that positive parenting behaviors may protect mental health in part by fostering more favorable self-concept development.

By contrast, maladaptive parenting styles were associated with increased mental health symptoms through their detrimental impact on self-concept. Significant indirect effects through self-concept were observed for paternal punishment and severity (indirect effect, 0.0007; mediation proportion, 84.89%), paternal rejection and denial (indirect effect, 0.0013; mediation proportion, 81.96%), paternal overprotection (indirect effect, 0.0007; mediation proportion, 56.24%), maternal punishment and severity (indirect effect, 0.0011; mediation proportion, 89.44%), maternal rejection and denial (indirect effect, 0.0011; mediation proportion, 88.23%), and maternal overinterference and overprotection (indirect effect, 0.0001; mediation proportion, 71.64%; all P < 0.001). No significant mediation via self-concept was found for parental preference for the child.

Family functioning also significantly mediated the associations between parenting styles and mental health symptoms, although these effects were generally smaller than those for self-concept. Paternal and maternal emotional warmth were indirectly associated with fewer mental health symptoms through better family functioning, with indirect effects of -0.0057 and -0.0061 and mediation proportions of 42.78% and 55.8%, respectively (both P < 0.001). These findings indicate that emotionally supportive parenting may benefit psychological outcomes by promoting healthier family environments.

Conversely, maladaptive parenting styles were linked to elevated mental health symptoms via impaired family functioning. Significant indirect effects through family functioning were observed for paternal punishment and severity (indirect effect, 0.0003; mediation proportion, 29.22%), paternal rejection and denial (indirect effect, 0.0005; mediation proportion, 26.19%), paternal overprotection (indirect effect, 0.0003; mediation proportion, 18.32%), maternal punishment and severity (indirect effect, 0.0005; mediation proportion, 32.1%), maternal rejection and denial (indirect effect, 0.0003; mediation proportion, 25.39%), and maternal overinterference and overprotection (indirect effect, 0.0001; mediation proportion, 21.09%; all P < 0.001). Again, no significant mediation via family functioning was observed for parental preference for the child.

These results indicate that both self-concept and family functioning are important psychological and familial mechanisms through which parenting styles influence mental health outcomes in children and adolescents.

DISCUSSION

This large cross-sectional study of 4433 Chinese children and adolescents adds to the understanding of how parenting styles, self-concept, and family functioning interrelate with mental health symptoms. Three key findings were identified. First, emotional warmth from both fathers and mothers was significantly associated with a lower risk of mental health symptoms, whereas maladaptive parenting practices (such as punishment, rejection, and overprotection) were associated with increased psychological vulnerability. Second, both self-concept and family functioning independently mediated the associations between parenting styles and mental health symptoms, indicating that parenting behaviors may be associated with adolescent psychological well-being through both intrapersonal (self-concept) and familial (family functioning) pathways. Third, these mediation effects remained robust after extensive adjustment for sociodemographic confounders, supporting the stability of the observed relationships.

These findings align with a substantial body of literature highlighting the protective role of positive parenting in adolescent mental health[3,5]. Parker[16] proposed that parental bonding characterized by emotional warmth and developmentally appropriate autonomy forms a key basis for healthy psychological development. Conversely, harsh, punitive, and rejecting parenting practices have consistently been linked to elevated risks of both internalizing and externalizing problems[7,17]. Extending prior work, this study concurrently examined paternal and maternal parenting styles in a large Chinese sample and showed that emotional warmth from both parents was similarly associated with reduced odds of mental health symptoms. This highlights the need to consider the complementary contributions of both fathers and mothers in shaping adolescent psychological outcomes.

The mediating role of self-concept observed here is consistent with Roger’s humanistic theory[11], which posits that unconditional positive regard from caregivers supports the development of a coherent and integrated self, thereby enhancing resilience and emotional well-being[18]. In line with previous evidence[19,20], supportive parenting was associated with more positive self-concept, which in turn was linked to fewer mental health symptoms. Notably, the results also showed that self-concept mediated not only the beneficial effects of positive parenting but also the detrimental effects of maladaptive parenting, particularly rejection and overprotection. This suggests that negative parenting experiences may contribute to poorer psychological adjustment partly by eroding self-worth and self-acceptance.

A particularly noteworthy result is the suppression effect observed for emotional warmth. Mediation proportions exceeded 100% (121.48% paternal, 121.15% maternal), indicating inconsistent mediation: The indirect effect via self-concept was protective (-0.0131), whereas the direct effect was adverse (+0.0023). The total effect (-0.0108) remained protective because the indirect benefit outweighed the opposing direct component, a suppression pattern well documented in prior methodological work[15]. Two culturally specific mechanisms may account for this divergence between direct and indirect effects in Chinese families. First, emotional warmth in Chinese parenting is seldom expressed independently of high expectations and behavioral control. The traditional concept of “guan” integrates affection with academic surveillance, behavioral discipline, and conditional regard[21]. The self-concept pathway primarily reflects the genuine benefit of perceived affection, which was associated with reduced symptom risk, whereas the positive direct effect may capture residual elements of intensive parental involvement - such as performance pressure and constrained autonomy - that are not offset by self-concept and may independently contribute to distress. Second, self-concept functions as a pathway-specific mediator. Higher self-concept was linked to reduced distress through more favorable self-appraisal but does not necessarily attenuate external stressors, such as academic pressure, that co-occur with high-involvement parenting in Chinese contexts. Notably, this suppression pattern has not been documented in earlier work. A meta-analysis by Tehrani et al[22] synthesizing 53 studies identified partial mediation of parenting-depression associations by self-esteem, with all mediation proportions below 100%. Ma et al[23] likewise reported no suppression effects among Chinese adolescents. This discrepancy may arise from construct differences: Prior studies typically focused on parental autonomy support or Western-derived parenting dimensions, whereas the EMBU emotional warmth scale used in the present study may capture a broader composite in which warmth is tightly coupled with demandingness, aligning more closely with the indigenous construct of “guan”.

Family functioning also emerged as a key mediating mechanism linking parenting styles to adolescent mental health. Family environments characterized by effective communication, competent problem-solving, emotional responsiveness, and supportive relationships appear to provide a psychologically secure and emotionally nurturing context, which was associated with lower vulnerability to psychological distress[9,24]. Consistent with the family systems perspective articulated by Matejevic et al[8] the present findings suggest that parenting styles and family functioning operate as mutually embedded components within a larger relational system. Family functioning accounted for approximately 19%-54% of the total effects of parenting styles on mental health symptoms, highlighting the substantial influence of broader family dynamics. These results imply that interventions targeting family-level functioning may be as crucial as those focused on specific parenting behaviors.

Parental favoritism was not significantly associated with mental health symptoms, either directly or indirectly, a pattern that contrasts with prior evidence linking favoritism to sibling conflict, emotional maladjustment, and impaired psychosocial functioning[25]. Several factors may account for this discrepancy. First, a recent network meta-analysis[26] indicated that although parental differential treatment (PDT) is generally associated with child psychopathology, this association is attenuated in collectivistic cultures, which may help explain the null effect in this Chinese sample. Second, the EMBU preference subscale assesses global perceived parental preference rather than sibling-comparative differential treatment, which is typically captured by specialized PDT instruments in earlier work; in Chinese families, where differential treatment may be culturally normative, this broader measure may tap a construct with weaker psychological salience. Third, many prior studies reporting significant favoritism effects were based on clinical samples or Western populations and did not always adjust for correlated parenting dimensions such as emotional warmth, rejection, and punishment. Once these covariates were comprehensively controlled in the present analysis, the remaining independent contribution of favoritism was minimal. Future research using culturally adapted PDT measures and longitudinal designs is needed to clarify whether, and under what contextual conditions, parental favoritism affects child mental health in contemporary Chinese populations.

Several limitations should be acknowledged. First, all variables were assessed at a single time point, precluding causal inference and leaving the possibility of bidirectional or reverse associations unresolved. Second, the sampling strategy relied on a combination of online investigator recruitment, purposive school selection, and classroom-based enrollment. Adolescents who were out of school or absent on the survey day were not captured, potentially omitting youth with more severe difficulties or unstable family environments[27]. In addition, approximately 69.5% of participants were drawn from Shaanxi Province, constraining geographic representativeness. Probability-based sampling across a broader range of provinces would enhance generalizability. Third, 1188 participants (21.1%) were excluded because their MHT validity score was ≥ 7, indicating potentially unreliable responding. Excluded adolescents reported more adverse parenting experiences and higher symptom levels, although demographic characteristics were comparable (Supplementary Table 2). Inattentive respondents often over-endorse symptoms[28]. Yet elevations on validity scales can also appear in genuinely distressed individuals[29]. The decision to exclude these cases likely produced conservative, rather than inflated, estimates. Future studies would benefit from sensitivity analyses comparing models with and without validity-based exclusions to quantify possible selection bias. Fourth, reliance on adolescent self-report for all constructs may have introduced common method variance and reporting bias despite the use of validated scales.

Several strengths also enhance confidence in the findings. The relatively large sample and adjustment for multiple sociodemographic covariates support the robustness of the estimates. Use of widely applied instruments (EMBU, FAD, PHCSS, MHT) facilitates comparison with prior work. Joint examination of self-concept and family functioning yields a more integrated account of the intrapersonal and familial processes through which parenting styles may influence adolescent mental health.

CONCLUSION

Overall, the findings suggest that parenting styles are associated with adolescent mental health through self-concept and family functioning as mediating mechanisms. Because both self-concept and family functioning are modifiable through targeted psychosocial interventions, this finding may be important for clinicians to consider for adolescent mental health prevention and treatment strategies.