Maternal Psychopathology and Family Functioning as Predictors of Externalizing Behavior in Adolescents: A Cross-Sectional Study in Greece

Abstract

Adolescent externalizing problems are commonly linked to maternal psychological distress and family functioning, but these associations remain underexplored in the Greek sociocultural context. This study examined how maternal symptoms of depression, anxiety, and stress, along with adolescent-perceived family functioning, predict externalizing behaviors in Greek adolescents. A total of 563 adolescent–mother dyads (63.4% girls; M_age = 15.03 and SD = 0.83) participated. Mothers completed the Child Behavior Checklist (CBCL), the Depression Anxiety Stress Scales (DASS-21), and the Family Assessment Device (FAD–GF), while adolescents completed the Youth Self-Report (YSR) and FAD–GF. Hierarchical regression analysis showed that adolescent-perceived family functioning was the strongest predictor of externalizing behavior (β = 0.24 and p < 0.001), even after accounting for demographic and maternal mental health variables. The final model explained 18% of the variance in adolescent externalizing problems. Mediation analysis confirmed that family functioning partially mediated the relationship between maternal depression and adolescent externalizing problems, with a significant indirect effect (a × b = 0.088, Sobel z = 2.90, and p = 0.004). Gender differences were found for self-reported aggressive behavior (t = −2.40, p = 0.017, and d = 0.20), with girls scoring higher than boys. These findings highlight the indirect impact of maternal depression through family dynamics and underscore the importance of culturally sensitive, family-centered interventions to reduce adolescent externalizing problems.

1. Introduction

Adolescence is a transformative period characterized by rapid biological, cognitive, and social changes that not only shape identity and autonomy but also set the stage for future psychosocial functioning [1,2]. Although the conceptualization of adolescence has evolved over time, with modern definitions generally encompassing ages 10 to 19 [3,4], this stage remains inherently complex due to the interplay of cultural, developmental, and contextual influences [5]. Recognizing this complexity is essential for understanding both the opportunities for growth and the vulnerabilities that emerge during this period.

During adolescence, profound neurobiological reorganization occurs, particularly in brain regions associated with executive functioning, decision-making, and emotional regulation [6,7]. Such neural maturation supports the development of higher-order cognitive abilities but also contributes to increased impulsivity and susceptibility to risk-taking behaviors [8,9]. Concurrently, adolescents experience significant shifts in social roles and family dynamics as they negotiate increasing independence while still relying on familial support [10].

This dual process of separation and connection can give rise to behavioral challenges that manifest in two primary domains: internalizing behaviors (such as anxiety, depression, and social withdrawal) and externalizing behaviors (including aggression, defiance, and rule-breaking) [11,12,13]. In particular, externalizing behaviors are of considerable concern, as they are often linked with long-term social maladjustment, criminality, and difficulties in occupational functioning [14,15,16,17,18].

Although broad parental psychopathology has been linked to adolescent adjustment, mothers often serve as the primary attachment figure and day-to-day emotional regulators in the family [19,20]. Hence, maternal symptoms of depression, anxiety, and stress may exert particularly strong influences on adolescents’ behavior, both directly, through the modeling of maladaptive emotion regulation strategies, and indirectly by disrupting the family’s emotional climate. Examining maternal psychopathology in isolation allows us to (a) clarify its unique contribution beyond general parental distress and (b) test mechanistic pathways, such as impaired family functioning, through which maternal symptoms translate into externalizing problems in youths.

1.1. Maternal Psychopathology and Family Functioning

A substantial body of research has focused on the role of parental mental health in shaping adolescent behavioral outcomes, with maternal psychopathology emerging as a key predictor [19,20]. Mothers, frequently the primary caregivers, play a central role in the daily emotional and behavioral regulation of their children. When mothers experience elevated levels of depression, anxiety, or stress, these symptoms can interfere with their ability to provide consistent emotional support and effective discipline [21,22]. For example, maternal depression is often associated with less responsive and more punitive parenting styles, which can disrupt the development of secure attachment and adaptive coping strategies in adolescents [23,24]. Furthermore, the transmission of emotional distress may occur through both genetic predispositions and environmental interactions, leading adolescents to either internalize negative emotions or express them through disruptive, externalizing behaviors [25,26].

Family functioning constitutes another critical component in the developmental context of adolescence. The quality of familial interactions, including communication patterns, emotional connectedness, and conflict resolution strategies, can either buffer or exacerbate the effects of maternal psychopathology on adolescent behavior [27,28,29]. High levels of family cohesion and adaptive functioning have been shown to mitigate the negative impact of parental distress, whereas dysfunctional family environments characterized by high conflict and poor communication may amplify the risk of behavioral problems in adolescents [30,31]. Theoretical frameworks, such as family systems theory, underscore the importance of viewing individual behavior within the context of the entire family unit, as the well-being of one member is interdependent with the functioning of the whole system [32,33,34,35]. This perspective is particularly valuable in understanding how maternal mental health and overall family dynamics interact to influence adolescent outcomes.

The association between parental psychopathology, family functioning, and adolescent behavior can be understood through the framework of intergenerational transmission of psychopathology, which refers to the processes by which emotional and behavioral patterns are passed from parents to children through both genetic and environmental pathways [36]. Environmental influences include parenting practices, parent–child interaction quality, and the overall emotional climate of the household. Maternal psychological distress may contribute to adolescent externalizing problems, both directly, through impaired regulation and responsiveness, and indirectly, via disrupted family functioning or heightened relational stress. This perspective supports the examination of both maternal mental health and adolescents’ perceptions of family dynamics as interrelated predictors of behavioral outcomes. These interrelated processes support a mediational model, wherein maternal psychological distress contributes to adolescent externalizing problems through its impact on family functioning.

1.2. Cultural Context

Despite extensive research in predominantly Western contexts, there is a notable gap in studies that examine these complex relationships within culturally specific settings. In Greece, limited attention has been paid to the joint impact of maternal psychopathology and family functioning on adolescent externalizing behaviors [37,38,39,40]. Addressing this gap is crucial for developing contextually relevant, family-centered interventions.

Furthermore, previous research in Greece has emphasized the emotional interdependence between mothers and their adolescent children. For instance, Giannouli and Stoyanova [41] found that depressive symptoms in both teenage patients and their mothers were associated with lower satisfaction in doctor–patient communication, underscoring the psychological synchrony within mother–child dyads. Their findings also suggest that maternal emotional states can shape how adolescents engage with healthcare professionals and, potentially, other social systems. Greece’s cultural emphasis on familial interdependence may influence both the expression of adolescent behavioral problems and the impact of parental mental health. Such evidence further supports the importance of exploring maternal psychopathology and family functioning when investigating adolescent behavioral outcomes within the Greek sociocultural context. Understanding how these variables interact in Greek families can inform more culturally responsive prevention and intervention strategies.

The present study examines the interrelationships among maternal psychological distress, adolescent externalizing behavior, and family functioning in a sample of Greek adolescents and their mothers. Specifically, it seeks to address the following objectives: (1) assess the prevalence and severity of maternal depressive, anxiety, and stress-related symptoms and their associations with adolescent externalizing behaviors; (2) evaluate whether adolescent-perceived family functioning serves as a mediating mechanism in the link between maternal psychopathology and externalizing symptoms; (3) explore potential gender differences in externalizing behaviors and in the impact of maternal- and family-level factors; and (4) identify familial and contextual risk indicators that may increase the likelihood of persistent behavioral difficulties during adolescence.

By integrating frameworks from developmental psychopathology and the intergenerational transmission of risk, this study aims to provide a more nuanced understanding of the pathways through which maternal distress and family relational processes shape adolescent behavioral outcomes. These insights are particularly valuable within the Greek cultural context, where familial interconnectedness plays a central role in youth development. Findings may inform culturally sensitive intervention strategies and public health efforts aimed at supporting adolescents and their families.

1.3. Research Question and Hypotheses

The primary research question guiding this study is as follows: to what extent and by what mechanisms do maternal psychological distress and family functioning predict externalizing behavior problems in Greek adolescents, and does family functioning mediate this relationship?

Based on prior empirical findings and theoretical models, we hypothesize the following: (1) higher levels of maternal depression, anxiety, and stress will be positively associated with adolescent externalizing problems; (2) poorer family functioning, as perceived by adolescents, will also predict greater externalizing behaviors; (3) family functioning will mediate the relationship between maternal psychological distress and adolescent externalizing problems such that higher maternal distress will be linked to worse family functioning, which in turn will predict higher adolescent externalizing symptoms; and (4) gender differences may emerge in the manifestation of externalizing behaviors and in the strength of associations with maternal and family variables.

2. Materials and Methods

2.1. Study Design

This non-interventional, cross-sectional study was designed to investigate the relationship between adolescent externalizing behavior problems, maternal mental health difficulties (specifically depression, anxiety, and stress), and family functioning. Data were collected using convenience sampling. The study targeted secondary school students (aged 11–17 years) from public middle and high schools in the Attica region and their biological mothers.

2.2. Participants

The target sample comprised adolescents enrolled in the second and third grades of middle school and the first grade of high school (Grades 8, 9, and 10), along with their mothers. Inclusion criteria for adolescents were enrollment in one of the specified grades, adequate proficiency in understanding and communicating in Greek (both orally and in writing), and the provision of parental consent. An initial list of 38 candidate schools in the Attica region was compiled, and invitations to participate were sent to these schools. Ultimately, 18 schools consented to take part. These schools were distributed across various sub-regions, including eight from Central Athens, eight from the southern sector, two from East Attica, and two from Piraeus.

Inclusion criteria for adolescents were (a) enrollment in the Grade 8, 9, or 10; (b) ages between 11 and 17 years; (c) adequate proficiency in Greek (both oral and written); and (d) provision of informed parental consent. For mothers, inclusion criteria were (a) being the biological mother of the participating adolescent and (b) the completion of all required questionnaires. Exclusion criteria included the following: (a) adolescents or mothers with incomplete data on key variables, (b) non-biological maternal caregivers, and (c) families where the adolescent had a severe intellectual disability or a diagnosed neurodevelopmental disorder that would interfere with questionnaire completion.

Of the eligible participants, approximately 30% ultimately consented to participate. This participation rate reflects opt-in consent procedures commonly used in school-based psychological research in Greece and may have been influenced by logistical challenges, including parental availability, exam periods, and absenteeism on the day of data collection. Although this rate introduces potential concerns about selection bias and generalizability, the final sample was demographically diverse and reflected a broad range of socioeconomic backgrounds, educational levels, and family structures across multiple districts in the Attica region.

2.3. Procedure

Data were collected from December 2022 to April 2023, following approval from the Institute of Educational Policy, the Regional Directorate of Secondary Education, and the Ethics Committee of the School of Medicine of the National and Kapodistrian University of Athens. Participating schools were contacted by the research team, and school principals distributed study information and parental consent forms via email. During class hours, students received a brief in-person explanation of the study. Those with parental consent completed questionnaires in class under supervision. Each student also delivered a questionnaire pack to their mother, who completed and returned it through the student. Student participation lasted approximately 25 min and included a brief Q&A session.

2.4. Measures

2.4.1. Demographic Questionnaire

Parents and adolescents completed a custom demographic questionnaire. Parent items included age, education, employment, socioeconomic status (SES), health status, and family structure. Adolescents reported their age, gender, school grade, perceived SES, and family composition.

2.4.2. Assessment of Behavioral Problems

Two standardized instruments were used to assess behavioral problems in adolescents:

• Youth Self Report (YSR): Developed by Achenbach [42], the YSR is a self-report instrument comprising 118 closed-ended items rated on a 3-point Likert scale (0 = “not true”, 1 = “somewhat or sometimes true”, and 2 = “very true or often true”). It evaluates eight syndromic scales (e.g., withdrawal, somatic complaints, anxiety/depression, social problems, thought problems, attention problems, delinquent behavior, and aggressive behavior) and yields two broad indices—internalizing and externalizing problems. The instrument classifies scores based on percentile cut-offs, where scores above the 98th percentile indicate clinical significance and those above the 90th percentile indicate a high risk for clinical problems. The YSR has been validated and adapted for Greek adolescents [43], and in this study, the internal consistency (Cronbach’s alpha) for the scales assessing aggressive behavior, rule-breaking, and overall externalizing problems ranged from 0.76 to 0.87.
• Child Behavior Checklist (CBCL) for Ages 6–18: Also developed by Achenbach [42], the CBCL is a parent-report instrument consisting of 113 items rated on the same 3-point Likert scale. It assesses the same eight syndromic scales and the broader internalizing and externalizing domains as the YSR. Adapted for the Greek population [44], the CBCL in this study demonstrated acceptable reliability, with Cronbach’s alpha values ranging from 0.77 to 0.90 for the scales used.

2.4.3. Assessment of Maternal Mental Health

• Depression Anxiety Stress Scale—21 (DASS-21): The DASS-21 [45] comprises 21 items that measure three dimensions: depression, anxiety, and stress (7 items each). Respondents rate the frequency of their symptoms over the past week on a 4-point Likert scale (0 = “Did not apply to me at all” to 3 = “Applied very much or most of the time”). Subscale scores range from 0 to 21, with higher scores indicating greater symptom severity. Specific cut-off scores are used to identify clinically significant levels (e.g., scores above 14 for depression, 10 for anxiety, and 17 for stress are indicative of severe symptoms). The DASS has been translated and validated in Greek [46], and in the current study, the internal consistency for the depression, anxiety, and stress subscales was 0.85, 0.83, and 0.88, respectively.

2.4.4. Assessment of Family Functioning

• General Functioning Scale of the Family Assessment Device (FAD–GF): The FAD–GF [47] consists of 12 self-report items designed to assess overall family functioning. Items are rated on a 4-point Likert scale ranging from 1 (strongly agree) to 4 (strongly disagree), with reverse scoring applied to certain items so that higher overall scores indicate lower family functioning. A score of 24 or higher is used to indicate poor family functioning. The FAD–GF has been adapted for the Greek context, demonstrating acceptable psychometric properties [48]. In the present study, internal consistency was excellent, with a Cronbach’s alpha of 0.90 for adolescent reports and 0.89 for parent reports. Although both adolescents and mothers completed the FAD–GF, only the adolescent-reported scores were included in the regression and mediation analyses. This decision was based on prior research suggesting that adolescents’ own perceptions of family functioning are more directly associated with their behavioral outcomes, particularly externalizing problems [49,50].

2.5. Ethical Considerations

The study was conducted in accordance with ethical standards and received approval from the Committee of Ethics and Deontology of the School of Medicine of the National and Kapodistrian University of Athens (protocol no. 106/27-06-2019). Informed consent was obtained from all parents/guardians prior to participation, and all participants (adolescents and their mothers) were assured that their involvement was voluntary and that they could withdraw at any time without penalty. Confidentiality was strictly maintained by ensuring that no personally identifiable information (such as names) was collected. Data were stored securely, and all procedures adhered to guidelines for the protection of personal data.

2.6. Statistical Analysis

Data analysis was performed using the SPSS software (version 29.0). Quantitative variables were described using means, standard deviations, and minimum and maximum values, while qualitative variables were summarized using absolute (N) and relative (%) frequencies. The normality of quantitative data distributions was assessed using the Kolmogorov–Smirnov test.

For inferential statistics, comparisons between two categorical variables were conducted using Pearson’s chi-square test or Fisher’s exact test when assumptions were not met. For the comparison of quantitative variables between two groups, either Student’s t-test or the Mann–Whitney U test was employed, depending on the distribution of the data. For comparisons involving more than two groups, one-way analysis of variance (ANOVA) was used. Effect sizes were computed for all inferential tests: Cohen’s d for t-tests, η² (eta squared) for ANOVA, and Cramér’s V for chi-square tests, as appropriate. Effect sizes were interpreted according to conventional benchmarks: small (d ≈ 0.20 and η² ≈ 0.01), medium (d ≈ 0.50 and η² ≈ 0.06), and large (d ≥ 0.80 and η² ≥ 0.14).

Multiple regression analysis (enter method) was conducted to determine the predictive value of maternal mental health and family functioning on adolescent externalizing behavior. Hierarchical regression was selected to examine the additive predictive value of conceptually distinct blocks of variables. In Step 1, demographic and contextual variables (e.g., adolescent and maternal age, maternal education and employment status, socioeconomic status, and health conditions) were entered to account for background characteristics. In Step 2, maternal psychological distress (DASS-21 subscales) was added, representing proximal intrapersonal factors. Finally, in Step 3, adolescent-reported family functioning (FAD–GF) was introduced as the most immediate relational factor. This order reflects a theoretically grounded progression from distal sociodemographic influences to proximal psychological and relational predictors of adolescent externalizing behavior, consistent with ecological and developmental models. Maternal DASS-21 depression, anxiety, and stress scores were treated as continuous variables in all analyses to preserve the dimensional nature of psychological symptom severity. Relationships between two quantitative variables were assessed using Pearson’s r or Spearman’s rs correlation coefficients. Correlation strengths were interpreted as low (r or rs between 0.10 and 0.30), moderate (between 0.31 and 0.50), or high (greater than 0.50). The internal consistency of the instruments was evaluated using Cronbach’s alpha.

In addition, a mediation analysis was conducted to examine whether adolescent-perceived family functioning mediated the relationship between maternal depression and adolescent externalizing behavior. The mediation model comprised paths a (independent variable → mediator), b (mediator → dependent variable), c (total effect), and c′ (direct effect). The significance of the indirect effect (a × b) was tested using the Sobel test.

Missing values in quantitative variables were replaced with the median of the respective variable. Only participants who met the age criteria (between 11 and 17 years) and whose questionnaires were completed by biological mothers were included in the final analyses. Median imputation was selected due to the relatively low rate of missingness (generally under 5% for any given variable) and to preserve the sample size for multivariate analyses. This method was preferred over mean substitution to reduce the influence of outliers, particularly for variables with non-normal distributions. Although more complex methods such as multiple imputation are recommended when missingness is substantial or systematic, median imputation was considered appropriate and unlikely to bias the results in the context of this dataset. All statistical tests were two-tailed, and the level of significance was set at 5%.

3. Results

3.1. Descriptive Statistics

The final sample consisted of 563 adolescent–mother dyads. As shown in

Table 1. Demographic characteristics of participants.

Variable	Total (n)	%	Girls (n)	%	Boys (n)	%	Statistical Test (p-Value)	Effect Size
Adolescent’s Age (years)	563	—	357	—	205 *	—	t = −1.23 (p = 0.220)	d = 0.10
Mean (SD)	15.03 (0.83)	—	15.06 (0.79)	—	14.98 (0.86)	—
School Grade (valid n = 537)							χ2 = 4.75 (p = 0.093)	V = 0.09
Grade 8	168	31.3%	98	28.7%	70	35.7%
Grade 9	190	35.4%	125	36.7%	65	33.2%
Grade 10	179	33.3%	118	34.6%	61	31.1%
Adolescent’s Nationality (valid n = 444)							–
Greek	436	98.2%	283	98.3%	152	98.1%
Albanian, Bulgarian or Middle Eastern	8	1.8%	5	1.7%	3	1.9%
Family Structure (valid n = 429)							χ2 = 0.46 (p = 0.496)	V = 0.03
Both parents	367	85.5%	240	86.0%	126	84.6%
Single parent	62	14.5%	39	14.0%	23	15.4%
Adolescent’s Health Problem (valid n = 562)	41	7.3%	29	5.2%	12	2.1%	χ2 = 0.81 (p = 0.368)	V = 0.04
Mother’s Age (years)							t = 0.81 (p = 0.419)	d = −0.06
Mean (SD)	47.64 (4.93)	—	47.54 (4.97)	—	47.84 (4.87)	—
Mother’s Education (valid n = 560)							χ2 = 4.73 (p = 0.094)	V = 0.09
Primary	7	1.3%	6	1.7%	1	0.5%
Secondary	199	35.5%	117	33.0%	82	40.0%
Tertiary	354	63.2%	232	65.4%	122	59.5%
Mother’s Nationality (valid n = 443)							–
Greek	379	85.5%	242	84.0%	136	88.3%
Albanian, Bulgarian, or Middle Eastern	64	14.5%	46	16.0%	18	11.7%
Family SES (valid n = 559)							χ2 = 1.16 (p = 0.560)	V = 0.05
Low	57	10.2%	33	9.3%	24	11.8%
Medium	476	85.2%	306	86.2%	169	83.3%
High	26	4.7%	16	4.5%	10	4.9%
Employment Status (valid n = 564)							χ2 = 9.36 (p = 0.009)	V = 0.13
Employed	468	83.0%	291	81.7%	177	85.1%
Unemployed	60	10.6%	36	10.1%	24	11.5%
Not Economically Active	36	6.4%	29	8.1%	7	3.4%
Mother Health Problem (valid n = 560)	45	8.0%	30	8.5%	15	7.3%	χ2 = 1.11 (p = 0.293)	V = 0.04

Note: Percentages are calculated based on the number of valid responses for each variable. For variables with missing data, the denominator (valid n) is indicated in the row label; * Boys column: n = 205; one adolescent’s gender was unreported. Effect sizes are reported as Cohen’s d for t-tests and Cramér’s V for chi-square tests. Values of Cohen’s d around 0.20 and Cramér’s V around 0.10 indicate small effects.

, adolescents’ ages ranged from 12 to 17 years (M = 15.03 and SD = 0.83), with 63.4% identifying as female (n = 357). No significant gender differences were found in age (t = −1.23, df = 560, p = 0.220, and Cohen’s d = 0.10).

The majority of adolescents were of Greek nationality (98.2%), with a small number identifying as having Albanian, Bulgarian, or Middle Eastern backgrounds. Most lived in two-parent households (85.5%). Around 7.3% reported having a health problem.

Mothers’ ages ranged from 26 to 70 years (M = 47.64 and SD = 4.93). Most had completed tertiary education (63.2%) and reported medium socioeconomic status (85.2%). While maternal employment status differed significantly by the child’s gender (χ² = 9.36, p = 0.009, and Cramér’s V = 0.13), all other demographic comparisons were non-significant, with small or negligible effect sizes.

3.2. Descriptive Statistics of Study Scales

The descriptive statistics for the study instruments are summarized in

Table 2. Descriptive statistics of study scales.

Instrument	Scale/Dimension	Total Sample Mean (SD)	Girls Mean (SD)	Boys Mean (SD)	t-Test (df)	Cohen’s d
CBCL	Aggressive Behavior	5.21 (4.81)	5.25 (4.61)	5.13 (5.17)	–0.28 (560)	0.02
	Rule-Breaking	2.19 (2.84)	2.03 (2.69)	2.46 (3.06)	1.73 (560)	–0.15
	Externalizing Problems	7.30 (7.15)	7.16 (6.81)	7.52 (7.73)	0.58 (560)	–0.05
YSR	Aggressive Behavior	7.32 (3.99)	7.62 (4.04)	6.79 (3.85)	–2.40 (560) *	0.20
	Rule-Breaking	3.99 (3.30)	3.84 (3.17)	4.24 (3.48)	1.41 (560)	–0.13
	Externalizing Problems	11.22 (6.43)	11.32 (6.34)	11.01 (6.62)	–0.55 (560)	0.03
DASS-21	Depression	3.83 (3.53)	3.75 (3.61)	3.93 (3.34)	0.61 (560)	–0.05
	Anxiety	1.69 (2.75)	1.76 (2.90)	1.55 (2.47)	–0.88 (560)	0.08
	Stress	2.53 (3.99)	2.52 (3.50)	2.54 (3.23)	0.05 (560)	0.01
FAD–GF	Mother’s Report	19.13 (5.60)	19.18 (5.62)	19.05 (5.56)	–0.25 (560)	0.02
	Adolescent’s Report	21.12 (6.17)	21.20 (6.74)	20.95 (5.04)	–0.51 (560) **	0.04

Note: * p < 0.05; ** p < 0.01. Cohen’s d values: small ≥ 0.20, medium ≥ 0.50, and large ≥ 0.80.

. Both parent-reported (CBCL) and adolescent self-reported (YSR) measures of externalizing behavior were utilized. Overall, the CBCL and YSR scores indicate moderate levels of externalizing problems. For example, the overall externalizing problems scores were 7.30 (SD = 7.15) for the CBCL and 11.22 (SD = 6.43) for the YSR, with the subscale scores for aggressive behavior and rule-breaking following similar patterns.

Maternal psychopathology, as measured by the DASS-21, reflected relatively low average levels of depression (M = 3.83 and SD = 3.53), anxiety (M = 1.69 and SD = 2.75), and stress (M = 2.53 and SD = 3.99) across the sample. In addition, the family functioning scores obtained from both mother (M = 19.13 and SD = 5.60) and adolescent (M = 21.12 and SD = 6.17) reports via the FAD–GF suggest that, on average, families exhibited satisfactory overall functioning.

Notably, while most comparisons between girls and boys did not reach statistical significance, a significant difference was observed on the YSR aggressive behavior subscale. Girls scored significantly higher than boys on this subscale (girls: M = 7.62 and SD = 4.04; boys: M = 6.79 and SD = 3.85; t = −2.40, df = 560, p < 0.05, Cohen’s d = 0.20, and a small effect), whereas no significant gender differences emerged for the other scales. No significant gender differences emerged on the other scales, and corresponding effect sizes were negligible to small: rule-breaking (d = −0.13), total YSR externalizing problems (d = 0.03), CBCL subscales (d range = −0.05 to 0.06), DASS-21 scores (d range = −0.10 to 0.06), and FAD-GF scores (d range = −0.03 to 0.04).

3.3. Group Comparisons by Externalizing Problem Classification

Based on the CBCL externalizing scale, adolescents were classified into three groups: normal, borderline, and clinical. Valid CBCL classification responses were available for 186 boys and 330 girls.

Among boys, 77.4% were classified as normal, 11.3% as borderline, and 11.3% as clinical. Statistically significant differences were observed across these groups in maternal education (χ² = 10.4, p = 0.034 and Cramér’s V = 0.17), child nationality (χ² = 10.5, p = 0.005, and V = 0.24), and the presence of health problems in the adolescent (χ² = 4.9, p = 0.087, V = 0.16, and a small-to-medium effect). Effect sizes suggest small-to-moderate practical significance (

Table 3. Demographic characteristics of boys by CBCL externalizing classification.

Variable	Total Boys (n, %)	Normal (n, %)	Borderline (n, %)	Clinical (n, %)	χ2 (p-Value)	Cramér’s V
School Grade					7.6 (p = 0.268)	0.14
– Grade 8	70 (34.1%)	52 (28.0%)	6 (3.2%)	6 (3.2%)
– Grade 9	65 (31.7%)	39 (21.0%)	7 (3.8%)	12 (6.5%)
– Grade 10	60 (29.3%)	46 (24.7%)	7 (3.8%)	3 (1.6%)
Mother’s Education					10.4 (p = 0.034)	0.17
– Primary	1 (0.5%)	0 (0.0%)	0 (0.0%)	1 (0.5%)
– Secondary	82 (40.0%)	54 (29.0%)	11 (5.9%)	9 (4.8%)
– Tertiary	121 (59.0%)	90 (48.4%)	10 (5.4%)	10 (5.4%)
Mother’s Nationality					1.5 (p = 0.480)	0.09
– Greek	136 (66.3%)	103 (55.4%)	13 (7.0%)	12 (6.5%)
– Other	18 (8.8%)	11 (5.9%)	2 (1.1%)	3 (1.6%)
Adolescent’s Nationality					10.5 (p = 0.005)	0.24
– Greek	152 (74.1%)	114 (61.3%)	15 (8.1%)	13 (7.0%)
– Other	3 (1.5%)	1 (0.5%)	0 (0.0%)	2 (1.1%)
Family Structure					2.2 (p = 0.330)	0.11
– Both parents	144 (70.2%)	94 (50.5%)	14 (7.5%)	12 (6.5%)
– Single parent	23 (11.2%)	16 (8.6%)	1 (0.5%)	3 (1.6%)
Adolescent’s Health Problem					4.9 (p = 0.087)	0.16
– Present	12 (5.9%)	6 (3.2%)	2 (1.1%)	3 (1.6%)
Mother’s Employment					6.9 (p = 0.225)	0.14
– Employed	176 (85.9%)	126 (67.7%)	15 (8.1%)	18 (9.7%)
– Unemployed	24 (11.7%)	15 (8.1%)	4 (2.2%)	3 (1.6%)
– Not Active	6 (2.9%)	3 (1.6%)	2 (1.1%)	0 (0.0%)

Note: Percentages in the “Total Boys” column are based on N = 205. Percentages in the “Normal”, “Borderline”, and “Clinical” columns are based on valid CBCL classification responses (n = 186). Effect sizes are presented as Cramér’s V; values ≥ 0.10 = small and ≥0.20 = medium.

).

Among girls, 78.8% were classified as normal, 7.9% as borderline, and 13.3% as clinical. Significant differences were found in maternal education (χ² = 15.8, p = 0.003, and V = 0.22), child nationality (χ² = 9.2, p = 0.010, and V = 0.17), family structure (χ² = 9.4, p = 0.009, and V = 0.17), and adolescent health problems (χ² = 7.5, p = 0.023, and V = 0.15). These effect sizes fall within the small-to-moderate range (

Table 4. Demographic characteristics of girls by CBCL externalizing classification.

Variable	Total Girls (n, %)	Normal (n, %)	Borderline (n, %)	Clinical (n, %)	χ2 (p-Value)	Cramér’s V
School Grade					3.4 (p = 0.492)	0.09
– Grade 8	98 (27.4%)	76 (21.3%)	5 (1.4%)	9 (2.5%)
– Grade 9	125 (35.0%)	92 (25.8%)	10 (2.8%)	13 (3.6%)
– Grade 10	118 (33.1%)	84 (23.5%)	9 (2.5%)	18 (5.0%)
Mother’s Education					15.8 (p = 0.003)	0.22
– Primary	6 (1.7%)	4 (1.1%)	0 (0.0%)	1 (0.3%)
– Secondary	117 (32.8%)	70 (19.6%)	16 (4.5%)	18 (5.0%)
– Tertiary	232 (65.0%)	185 (51.8%)	10 (2.8%)	24 (6.7%)
Mother’s Nationality					3.5 (p = 0.172)	0.10
– Greek	242 (67.8%)	188 (52.7%)	13 (3.6%)	28 (7.8%)
– Other	46 (12.9%)	30 (8.4%)	1 (0.3%)	9 (2.5%)
Adolescent’s Nationality					9.2 (p = 0.010)	0.17
– Greek	283 (79.3%)	216 (60.5%)	14 (3.9%)	34 (9.5%)
– Other	5 (1.4%)	2 (0.6%)	0 (0.0%)	3 (0.8%)
Family Structure					9.4 (p = 0.009)	0.17
– Both parents	240 (67.2%)	185 (51.8%)	9 (2.5%)	33 (9.2%)
– Single parent	39 (10.9%)	28 (7.8%)	3 (0.8%)	3 (0.8%)
Adolescent’s Health Problem					7.5 (p = 0.023)	0.15
– Present	29 (8.1%)	18 (5.0%)	4 (1.1%)	6 (1.7%)
Mother’s Employment					3.1 (p = 0.540)	0.09
– Employed	291 (81.5%)	213 (59.7%)	21 (5.9%)	34 (9.5%)
– Unemployed	36 (10.1%)	26 (7.3%)	4 (1.1%)	4 (1.1%)
– Not Active	29 (8.1%)	20 (7.7%)	1 (0.3%)	6 (1.7%)

Note: Percentages in the “Total Girls” column are based on N = 357. Percentages in the “Normal”, “Borderline”, and “Clinical” columns are calculated relative to the total sample of girls (N = 357); p-values represent chi-square test results for categorical comparisons. Effect sizes are presented as Cramér’s V; values ≥ 0.10 = small and ≥0.20 = medium.

).

3.4. Hierarchical Regression Analysis

To examine the predictive value of demographic characteristics, maternal mental health, and family functioning on adolescent externalizing problems (as measured by the YSR), a hierarchical regression analysis was conducted (see

Table 5. Hierarchical regression analysis predicting adolescent externalizing problems (YSR).

Model	Predictor	B	SE	Beta	p-Value	95% CI
Step 1	Constant	16.75	8.10	–	0.039	[0.80, 32.70]
	Adolescent’s Gender	0.66	0.74	0.04	0.372	[−0.79, 2.11]
	Adolescent’s Age	–0.17	0.43	–0.20	0.694	[−1.01, 0.68]
	Mother’s Age	–0.07	0.07	–0.05	0.295	[−0.20, 0.06]
	Secondary Education (Mother)	–3.50	3.36	–0.23	0.299	[−10.09, 3.09]
	Tertiary Education (Mother)	–5.98	3.34	–0.39	0.074	[−12.54, 0.57]
	Low SES	0.93	2.08	0.04	0.656	[−3.16, 5.02]
	Medium SES	1.24	1.78	0.06	0.486	[−2.25, 4.73]
	Mother’s Health Problem	2.93	1.24	0.11	0.019 *	[0.50, 5.37]
	Adolescent’s Health Problem	3.07	1.20	0.12	0.011 *	[0.72, 5.42]
	Mother Employed	3.33	1.64	0.17	0.042 *	[0.11, 6.54]
	Mother Unemployed	3.98	1.91	0.17	0.038 *	[0.24, 7.72]
	Step 1 Model (R2, Adj. R2)	0.27, 0.08
Step 2	Add: DASS-21—Depression	0.28	0.15	0.13	0.075	[−0.02, 0.58]
	Add: DASS-21—Anxiety	0.22	0.17	0.08	0.210	[−0.11, 0.55]
	Add: DASS-21—Stress	0.09	0.16	0.04	0.598	[−0.21, 0.39]
	Step 2 Model (R2, Adj. R2)	0.35, 0.12
Step 3	Add: FAD-GF	3.72	0.72	0.24	<0.001 **	[2.30, 5.14]
	Step 3 Model (R2, Adj. R2)	0.42, 0.18

Note: Standardized beta coefficients (β) are reported as indicators of effect size. Confidence intervals (95% CI) are provided for significant predictors in the final model. Adjusted R² reflects the proportion of variance in adolescent externalizing behavior explained by each model. * p < 0.05; ** p < 0.001.

). In Step 1, demographic variables (adolescent’s gender and age, mother’s age, and indicators of mother’s education, economic level, and employment status) accounted for 8% of the variance in externalizing problems (adjusted R² = 0.08). In this step, the presence of a health problem in the mother (β = 2.93 and p = 0.019) and in the adolescent (β = 3.07 and p = 0.011) as well as maternal employment status (employed: β = 3.33 and p = 0.042 and unemployed: β = 3.98 and p = 0.038) emerged as significant predictors.

In Step 2, maternal clinical characteristics (depression, anxiety, and stress scores) were added. The model’s explanatory power increased to 12% (Adjusted R² = 0.12), with maternal health problems and adolescent’s health problems remaining as significant predictors. However, the DASS-21 subscales did not significantly contribute to these indicators.

In Step 3, the family functioning score (adolescent report on the FAD-GF) was entered, resulting in a final model that explained 18% of the variance in externalizing problems (adjusted R² = 0.18). At this final step, maternal health problems (β = 2.45, p = 0.039, and 95% CI [0.13, 4.77]) and lower family functioning (β = 3.72, p < 0.001, and 95% CI [2.30, 5.14]) were significant predictors, indicating that, beyond demographics and maternal clinical status, family functioning significantly contributes to the prediction of externalizing behavior. Notably, maternal employment status and adolescent health status, which were significant in Step 1, no longer remained significant once maternal distress and family functioning were included in the model, suggesting that these proximal factors account for additional variance beyond demographic characteristics.

3.5. Correlation Analysis

Spearman’s rho (rs) correlations were conducted to examine the associations between adolescent externalizing behavior, maternal psychopathology, and family functioning (

Table 6. Spearman’s correlations among externalizing behavior, maternal psychopathology, and family functioning.

Variables	1	2	3	4	5
1. CBCL Externalizing	–
2. YSR Externalizing	0.40 **	–
3. DASS-21 Depression	0.36 **	0.29 **	–
4. FAD–GF (Adolescent)	0.28 **	0.35 **	0.39 **	–
5. Mother’s Education (ordinal)	–0.14 *	–0.09	–0.07	–0.04	–

Note: Spearman’s rho used for all tests. * p < 0.01; ** p < 0.001. All correlations are positive unless noted. Strength interpreted as small (0.10–0.30), moderate (0.31–0.50), and strong (>0.50).

). All correlations were statistically significant and ranged in strength from small to moderate.

A moderate positive correlation was observed between adolescent self-reported externalizing behavior on the YSR and parent-reported externalizing behavior on the CBCL (rs = 0.40 and p < 0.001), suggesting consistency between informants. The subscales of the DASS-21 (depression, anxiety, and stress) were also significantly intercorrelated, with coefficients ranging from rs = 0.54 to rs = 0.72 (p < 0.001 for all), indicating moderate to strong associations between dimensions of maternal psychological distress.

Maternal DASS-21 scores showed significant, positive associations with both adolescent and maternal reports of family dysfunction. Specifically, depression, anxiety, and stress were each positively correlated with the FAD–GF scores (rs range = 0.26 to 0.39, with p < 0.001 for all), indicating that higher psychological distress in mothers was associated with poorer family functioning.

Regarding externalizing behavior, the DASS-21 depression and anxiety scores were positively correlated with CBCL and YSR externalizing scores (rs = 0.26 to 0.37 and p < 0.001), reflecting small to moderate effects. FAD–GF scores, as reported by adolescents, were also moderately associated with externalizing behavior, particularly in the YSR ratings (rs = 0.35 and p < 0.001).

Finally, CBCL externalizing scores were negatively associated with maternal education level (rs = −0.14 and p = 0.001) and positively associated with the presence of health problems in both mothers (rs = 0.08 and p = 0.040) and adolescents (rs = 0.14 and p = 0.004), although these effects were relatively weak.

3.6. Mediation Analysis

To further examine the mechanisms linking maternal psychological distress and adolescent externalizing behavior, a mediation analysis was conducted to test whether adolescent-perceived family functioning served as a mediator. Maternal depression (DASS-21) was used as the independent variable, adolescent-reported family functioning (FAD–GF) as the mediator, and adolescent externalizing behavior (YSR) as the outcome (

Table 7. Mediation model testing family functioning as a mediator between maternal depression and adolescent externalizing behavior.

Path	B	SE	p-Value
a: Depression → Family Functioning	0.022	0.007	0.001
b: Family Functioning → Externalizing	4.05	0.91	<0.001
c: Total Effect (Depression → Externalizing)	0.51	0.12	<0.001
c′: Direct Effect (Depression → Externalizing, controlling for Family Functioning)	0.43	0.12	<0.001
Indirect Effect (a × b)	0.088	—	—
Sobel z	2.90	—	0.004

Note: All values are based on ordinary least squares regression models. The Sobel test was used to evaluate the significance of the indirect effect.

).

The analysis revealed that higher maternal depression was significantly associated with greater family dysfunction (B = 0.022 and p = 0.001; Path a), and greater family dysfunction was significantly associated with higher externalizing behavior, even after controlling for maternal depression (B = 4.05 and p < 0.001; Path b). The total effect of maternal depression on adolescent externalizing behavior was significant (B = 0.51 and p < 0.001; Path c), and the direct effect remained significant though reduced when family functioning was included (B = 0.43 and p < 0.001; Path c′), indicating partial mediation. The indirect effect (a × b = 0.088) was statistically significant according to the Sobel test (z = 2.90 and p = 0.004), supporting the mediating role of family functioning in this relationship.

4. Discussion

The present study aimed to investigate the manifestation of externalizing problems in adolescents in relation to maternal mental health issues and family functioning. Our findings indicate that adolescents’ externalizing behaviors are significantly associated with maternal depression, anxiety, and stress, with the strongest association observed between adolescent externalizing problems and maternal depression. In addition, lower levels of family functioning were found to be significantly related to higher levels of externalizing problems in adolescents. These findings provide important insights into the complex interplay between maternal well-being, family dynamics, and adolescent behavioral outcomes.

Our findings indicate that moderate maternal stress and depression are positively correlated with externalizing behaviors in adolescents. Specifically, mothers of children scoring in the borderline or clinical range on these measures reported significantly higher levels of depressive symptoms and stress. However, it is important to note that in the multivariate regression model, maternal depression, anxiety, and stress did not emerge as independent predictors once family functioning and other covariates were accounted for. This pattern supports previous research linking maternal psychological distress with adverse child outcomes and highlights the critical role maternal mental health may play in the development of externalizing behaviors [51,52,53], although the influence may be more indirect or mediated through broader family dynamics.

Furthermore, the present study demonstrates that family functioning plays a crucial role in moderating these associations. Lower family functioning, as indicated by less cohesive communication, reduced support, and inflexible family dynamics, was significantly associated with higher levels of externalizing behaviors. Mothers whose children were classified in the borderline or clinical ranges on externalizing problems reported poor family functioning. This supports the notion that a well-functioning family environment can serve as a protective buffer, potentially mitigating the negative effects of maternal distress on adolescent behavior [54,55].

Another important aspect that emerged from the data concerns the demographic characteristics of the mothers. Most of the participating mothers were employed, which may point to the challenge of balancing professional responsibilities with the demands of parenting. While our study specifically focused on maternal mental health due to its well-documented impact on child outcomes, it is important to recognize that fathers and other caregivers also contribute significantly to the emotional climate and functioning of the family system. In many modern societies, mothers continue to bear a disproportionate share of caregiving responsibilities, even when employed full-time. This dual burden can strain maternal resources and exacerbate stress levels [56]. Such parental stress may, in turn, compromise the quality of caregiving and reduce the availability of emotional support, thereby increasing the risk of externalizing behaviors in adolescents. These findings resonate with previous work emphasizing the pressures of managing multiple roles and their impact on family dynamics [57].

In addition, our analysis revealed that lower maternal education and the presence of health problems in the child were associated with higher externalizing behaviors. Lower educational attainment may limit a mother’s access to resources, knowledge, and effective parenting strategies, all of which are important for fostering a supportive home environment. This is in line with research suggesting that higher maternal education is linked to better parenting practices, improved socioeconomic conditions, and enhanced emotional regulation [58].

Our hierarchical regression analysis further illuminated the predictive power of these variables. When demographic factors were entered in the first step, they accounted for approximately 8% of the variance in adolescent externalizing problems. With the addition of maternal clinical characteristics (depression, anxiety, and stress) in the second step, the explained variance increased to 12%; however, these clinical indices did not significantly predict externalizing behavior beyond the effect of maternal and child health problems. In the final model, which included family functioning, the explained variance reached 18%. Here, maternal health problems and lower family functioning emerged as robust predictors, suggesting that interventions aimed at improving family dynamics might significantly reduce externalizing behaviors in adolescents.

In addition to the direct effects of maternal psychological distress and family dysfunction, the mediation analysis provided further insights into the mechanisms linking these factors. Specifically, adolescent-perceived family functioning partially mediated the association between maternal depression and adolescent externalizing behavior. This finding suggests that maternal symptoms may contribute to behavior problems not only through direct psychological influences but also by shaping the family’s emotional and relational climate. This aligns with intergenerational models of risk transmission [36] and underscores the importance of targeting family-level functioning, particularly as perceived by adolescents, in intervention efforts. Supporting family cohesion and communication may buffer the impact of maternal distress on adolescent adjustment.

4.1. Implications for Practice and Policy

The findings of this study have several important clinical and policy implications. First, they underscore the necessity for the early detection of maternal psychological distress and the implementation of interventions designed to alleviate maternal depression and stress. By addressing maternal mental health proactively, it may be possible to prevent or reduce the development of externalizing behaviors in adolescents. Second, the significant role of family functioning suggests that interventions aimed at enhancing family communication, cohesion, and adaptability could serve as effective strategies for mitigating behavioral problems in youths. Family-based interventions that foster supportive interactions and improve parental self-efficacy might be particularly beneficial.

Given that many mothers in our sample were employed, policies that support work–life balance, such as flexible work arrangements, accessible childcare, and family support programs, could also help reduce parental stress and promote healthier family environments. Encouraging workplaces to recognize and support the dual roles of working mothers may indirectly contribute to better mental health outcomes for both parents and children.

4.2. Limitations and Future Directions

Despite the valuable insights provided by our study, several limitations must be acknowledged. First, the sample was drawn exclusively from secondary schools in the Attica region, which may limit the generalizability of the findings to adolescents in other regions or settings. Future research should aim to include a more nationally representative sample. Although a stratified approach was used to recruit schools across various sub-regions (urban, suburban, and semi-rural), participation was voluntary, and school recruitment was affected by logistical constraints, such as exam periods and administrative approval timelines. Second, the relatively low response rate may affect the validity and reliability of the results and calls for cautious interpretation. While this response rate is typical of school-based research relying on opt-in parental consent in Greece, it introduces potential selection bias. Nonetheless, the final sample was demographically diverse and included families from a range of socioeconomic backgrounds, mitigating some concerns regarding representativeness.

Third, due to constraints in data collection, our study did not incorporate other potentially influential variables, such as peer influences, neighborhood safety, and the broader school environment, which could also affect adolescent externalizing behavior. Moreover, the explained variance of 18% in the final regression model, while statistically meaningful, underscores the complexity of adolescent externalizing behavior and the likely contribution of other unmeasured influences. Factors such as peer dynamics and peer contagion [59], individual cognitive styles and beliefs [60,61], and the broader school climate [62,63] are also known to significantly shape behavioral outcomes in adolescence. Future research should incorporate such variables to better capture the multifactorial nature of these behaviors and improve explanatory power. Additionally, several conceptually important factors were not included and warrant attention in future research. One key area involves the transactional, bidirectional associations between parenting stress and adolescent behavior problems, which have been shown to evolve over time and vary depending on the levels of maternal affection. A longitudinal study by Chiang and Bai [64] demonstrated that in families with high maternal affection, parenting stress and externalizing symptoms mutually reinforced one another, highlighting the need to consider relational closeness as a moderator. Moreover, the study by Xerxa et al. [65] emphasizes that most bidirectional effects between parent and child psychopathology are likely driven by within-rater associations and shared method variance, rather than true child-to-parent influence. These findings align with the psychometric work by Olino et al. [66], which found minimal bias in the maternal reports of child symptoms due to maternal psychopathology. Lastly, Giannouli [67] calls for multilevel, family-centered models, especially when working with adolescents presented to clinical services. Future longitudinal studies should adopt these more nuanced, multi-informant, and systemic approaches to clarify causal and reciprocal influences in the development of externalizing problems.

Lastly, the study focused exclusively on maternal variables and did not include fathers or other caregivers in the analysis. This limits our understanding of the broader family system, as paternal mental health and involvement may also play important roles in shaping adolescent behavioral outcomes [68,69]. Future research should adopt a more inclusive, multi-informant approach that considers the contributions of all caregivers.

In summary, our findings suggest that maternal mental health issues, particularly depression and stress, are significantly associated with higher levels of externalizing behaviors in adolescents, and lower family functioning further exacerbates this relationship. The mediation results provide additional support for the indirect role of family functioning in this association, suggesting a mechanism through which maternal distress may impact adolescents. These results highlight the importance of integrated interventions that focus not only on improving maternal psychological well-being but also on enhancing overall family functioning.

Future research should expand on these findings by employing nationally representative samples and longitudinal designs to better understand the stability and temporal dynamics of these relationships. Moreover, incorporating both quantitative and qualitative methods may offer deeper insights into the subjective experiences of families and the mechanisms through which maternal psychopathology influences adolescent behavior. It will also be critical to examine the potential moderating effects of cultural and socioeconomic factors, which may shape the manifestation of these issues differently across diverse contexts.

Overall, this study reinforces the idea that a holistic approach that addresses both individual and systemic factors is essential for reducing adolescent externalizing behaviors and promoting healthier developmental trajectories.

5. Conclusion

The present study contributes to the growing body of studies on adolescent behavioral development by highlighting the significant associations between maternal psychological distress, family functioning, and externalizing behaviors in adolescents within the Greek cultural context. The findings underscore the salience of adolescent-perceived family functioning as both a direct predictor and a partial mediator in the relationship between maternal depression and externalizing symptomatology. While maternal psychological distress, particularly depression and stress, was associated with increased adolescent behavioral difficulties, its influence appeared to be partially indirect, operating through disruptions in family functioning.

These results underscore the importance of conceptualizing adolescent behavioral outcomes within an ecological and systemic framework. Interventions aimed at reducing adolescent externalizing behaviors may benefit from a dual focus on improving maternal mental health and enhancing family-level processes such as communication, cohesion, and emotional responsiveness. Moreover, the findings support the utility of adopting a family-centered approach in both research and clinical practice, particularly in sociocultural contexts where familial interdependence plays a central developmental role.

Future research would benefit from longitudinal designs and the inclusion of additional familial-, contextual-, and individual-level variables, including paternal factors and peer influences, to further elucidate the complex, bidirectional pathways underlying adolescent externalizing behavior. Such efforts will be instrumental in informing the development of comprehensive, culturally sensitive interventions targeting adolescent mental health.