Grandparental involvement and preschoolers’ externalizing problem behaviors: the mediating roles of family functioning and emotion regulation

Abstract

Background

Grandparents represent an important component of child-rearing practices in many contemporary Chinese families. However, limited research has investigated how grandparent involvement is associated with externalizing problem behaviors in preschoolers within three-generation households (where a child lives with both parents and one or more grandparents). This study investigates the relationships among grandparental involvement, family functioning, preschoolers’ emotion regulation, and externalizing problem behaviors.

Methods

Data were collected from 802 preschoolers aged 3 to 6 years residing in three-generation households through questionnaires, including measures of grandparental involvement, family functioning, emotion regulation, and externalizing problem behaviors. Statistical analyses were conducted using SPSS, including correlation analysis, regression analysis, and path analysis with the bootstrap method (5,000 iterations) to examine the mediation hypothesis.

Results

Family functioning and preschoolers’ emotion regulation significantly mediated the relationship between grandparental involvement and preschoolers’ externalizing problem behaviors, with a chain mediating effect size of 5.78% and a total mediating effect size of 48.35%.

Conclusion

Family functioning and preschoolers’ emotion regulation function as two significant mediators in the association between grandparental involvement and preschoolers’ externalizing problem behaviors. It is recommended that interventions targeting preschoolers’ externalizing problem behaviors in Chinese three-generation families be implemented to promote positive grandparental involvement, enhance family functioning, and strengthen preschoolers’ emotion regulation abilities.

Introduction

With rapid economic growth and increasing life expectancy, family structures and childcare arrangements have experienced substantial changes [1–2]. In contemporary Chinese families, young parents often encounter considerable occupational pressure and extended working hours, which frequently restrict their capacity for full-time childcare [3]. As a result, grandparent involvement, defined as the degree to which grandparents engage in their grandchildren’s daily care, education, and development, has become a common strategy supporting child-rearing in three-generation households [4–7]. Although earlier studies frequently associated grandparent care with elevated risks of externalizing problem behaviors in preschoolers due to possible inconsistencies in discipline [8–11], family systems theory provides a more comprehensive perspective [12, 13]. This theory conceptualizes the family as an interconnected system in which the functioning of one member or subsystem is related to the functioning of the entire unit. From this perspective, grandparent involvement represents not merely a caregiving arrangement but also a meaningful familial resource. When appropriately integrated, it may be linked to enhanced family adaptability and stability, and consequently may exert protective effects on child development [14–17].

To clarify how grandparent involvement affects child outcomes, it is necessary to consider the crucial mediating role of family functioning. This construct includes emotional bonding, communication patterns, and the family’s ability to adjust to external stressors [13]. According to family systems theory, grandparent involvement operates as a systemic input. By sharing caregiving responsibilities and offering support, grandparents may reduce parental stress and facilitate a more harmonious family environment [18]. Such support is theorized to be associated with improved family functioning, potentially through strengthening cohesion and minimizing conflict [19]. A well-functioning family system provides a supportive ecological context for child socialization [20]. In contrast, if intergenerational involvement is associated with boundary ambiguity or conflict, family functioning may be undermined.

Within this familial context, emotion regulation emerges as a central mechanism underlying the development of externalizing problem behaviors [21]. Emotion regulation denotes the capacity to monitor and modulate emotional responses, which is essential for social adjustment [22–23]. Developmental research suggests that children primarily acquire these skills through family interactions, a process referred to as emotional socialization [24]. A positive family climate, characterized by warmth and effective communication, offers children the security and behavioral modeling necessary to manage their emotions effectively [24, 25]. Accordingly, stronger family functioning is considered to create an environment that supports the development of emotion regulation skills in preschoolers.

Based on these theoretical foundations, this study proposes a serial mediation model to clarify the mechanism linking intergenerational care with child behavior. Our study argues that grandparent involvement is associated with lower levels of preschoolers’ externalizing problem behaviors, and that this association may be sequentially mediated by family functioning and emotion regulation [26, 27]. By clarifying this systemic pathway, the present research provides a more comprehensive understanding of how macro-level family dynamics influence micro-level individual competencies and identifies specific intervention targets. From a practical perspective, these findings may inform the development of family-centered programs aimed at optimizing intergenerational coparenting and emotional socialization to promote positive child development in Chinese three-generation households.

Theoretical development and hypotheses

The relationship between grandparent involvement and preschool children’s externalizing problem behaviors

In traditional Chinese family structures, grandparents generally occupy a central role in family relationships, holding relatively high social status and authority. Wang noted that the three-generation family model in China has gradually transitioned from being grandparent-centered to grandchild-centered [28]. Grandparents often regard their grandchildren as their closest companions and actively engage in childcare by assisting with daily care [29], providing cognitive stimulation [30], establishing close emotional bonds [31], and attending to their moral development and behavior [32]. Overall, grandparent involvement provides not only material assistance but also expands preschool children’s cognitive perspectives through the transmission of traditional culture, thereby facilitating their understanding of social rules and behavioral norms. The period between ages 3 to 6 is critical for learning through imitation among preschool children [33], and grandparents’ guidance in daily behaviors is associated with fewer maladaptive behaviors, such as aggression and rule-breaking, in preschool children. Moreover, grandparents’ parenting styles often demonstrate greater warmth and tolerance [34], which may complement the comparatively stricter parenting approaches of younger parents. This complementary dynamic is associated with reduced emotional volatility in stressful contexts and with a lower risk of externalizing problem behaviors. Luo et al. reported that grandparent involvement significantly decreased the occurrence of externalizing problem behaviors in preschool children after controlling for parenting stress [12]. This finding highlights the role of grandparent involvement as a supplementary parenting strategy and suggests that active engagement by grandparents contributes to positive behavioral development in preschool children. Such evidence aligns with broader findings indicating that internal individual characteristics are associated with external behavioral patterns and interpersonal functioning [35]. Therefore, this study hypothesizes that grandparent involvement is negatively associated with the incidence of externalizing problem behaviors in preschool children.

• Hypothesis 1: Grandparental involvement is negatively associated with preschool children’s externalizing problem behaviors.

The mediating role of family functioning

Family functioning refers to the effectiveness of emotional bonds, family rules, communication among members, and the capacity to cope with external events. It can be divided into two specific dimensions: family closeness and family adaptability [36]. Family systems theory emphasizes that, as an integrated whole, interactions among family members exert a substantial influence on the functionality and stability of the family system [24]. When young parents are unable to fully engage in parenting responsibilities due to occupational pressure, the practical support provided by grandparents, such as childcare and sharing household duties, is associated with improved balance between work and family life [37]. The active participation of grandparents in parenting reflects their care and support for their adult children’s families and is associated with higher levels of family adaptability and closeness, as well as reduced tension and conflict among members [38]. Additionally, some studies indicate that family functioning is significantly positively associated with favorable developmental outcomes, including preschool children’s autonomy and academic performance, while significantly negatively predicting behavioral problems and peer interaction difficulties [36]. Accordingly, this study proposes that family functioning mediates the relationship between grandparent involvement and preschool children’s externalizing problem behaviors.

• Hypothesis 2: Family functioning plays a mediating role in the relationship between grandparental involvement and preschool children’s externalizing problem behaviors.

The mediating role of emotion regulation

Emotion regulation refers to the internal and external processes involved in adjusting, evaluating, and managing emotional responses to achieve desired goals, such as controlling emotional expression and effectively handling negative emotions [39]. The capacity to regulate emotions constitutes a central component of emotional competence. Furthermore, international research has highlighted that emotional intelligence functions as a critical internal driver of positive behavioral outcomes and adaptive social interactions [40]. A substantial body of early research on emotional development suggests that emotion regulation skills develop gradually into adulthood [41], with the early years of life representing a crucial period for this development [42]. The preschool stage is characterized by rapid growth in children’s emotion regulation abilities. Evidence indicates that the cognition, understanding, and organization of emotion regulation strategies are formed between ages 3 and 5 [43]. As language and cognitive capacities advance [44], preschool children begin to identify and manage their emotions, gradually acquiring effective strategies to regulate emotions in adaptive ways [45]. Family systems theory conceptualizes the family as an “emotional unit” in which emotional and behavioral patterns among members are interconnected [24]. In families with grandparent involvement, grandparents represent an important source of social support and play a significant role in preschool children’s emotion regulation. Due to time constraints, urban young parents frequently experience work-family conflict, which intensifies stress and anxiety, limits interaction with preschool children, and renders parental presence a limited resource. Reduced parent-child interaction is negatively associated with children’s emotion regulation [46]. In this context, grandparents may function as protective buffers, mitigating these adverse effects [47]. Specifically, grandparents’ extensive caregiving experience enables them to understand and respond effectively to preschool children’s negative emotions, provide emotional reassurance, and assist children in coping with frustration [48, 49], thereby strengthening emotion regulation abilities. Effective emotion regulation not only supports socio-emotional development and positive peer relationships but also helps restrain externalizing problem behaviors, such as aggression [50]. Moreover, Wen et al. reported that emotion regulation constitutes a central link between grandparent-grandchild relationships and social adaptation [51], whereby positive internal working models promote adaptive attribution patterns and reduce negative emotions and problem behaviors. Based on these considerations, the present study proposes that preschool children’s emotion regulation mediates the relationship between grandparental involvement and externalizing problem behaviors.

• Hypothesis 3: Emotion regulation mediates the relationship between grandparental involvement and externalizing problem behaviors in preschool children.

The chain-mediating roles of family functioning and children’s emotion regulation

According to family systems theory, child development is influenced by a sequence of processes extending from family dynamics to individual competencies [52, 53]. Within this hierarchical framework, grandparent involvement shapes the family environment, thereby establishing the immediate ecological context reflected in family functioning [54]. In this context, children’s emotion regulation develops through emotional socialization processes (e.g., modeling and coregulation) as a proximal individual mechanism that directly influences behavioral expression [55]. Thus, a theoretical pathway is proposed in which distal factors (grandparent involvement) indirectly affect child outcomes through their association with the proximal family context and subsequent individual developmental processes.

Specifically, grandparent involvement serves as a key resource for maintaining systemic stability. By sharing childcare responsibilities, grandparents may buffer family stress and improve the overall emotional climate, which is associated with more positive intergenerational coparenting and reduced harsh parenting practices [56, 57]. Such support contributes to household stability and enhances family functioning [58]. Within this improved family system, supportive grandparents broaden the child’s “emotion regulation network”. Grandparents often act as secondary attachment figures who provide additional emotional security and model constructive conflict resolution strategies, thereby offering children multiple opportunities for coregulation [25, 59]. This layered support facilitates the internalization of effective regulation strategies [24]. Ultimately, enhanced emotion regulation skills function as a proximal behavioral mechanism, enabling children to inhibit impulsive responses and manage negative affect more effectively, which directly decreases the incidence of externalizing problem behaviors [60]. Consequently, a serial mediation pathway is proposed in which grandparent involvement positively influences family functioning, which subsequently fosters emotion regulation and is ultimately associated with lower levels of externalizing problem behaviors [26, 27].

• Hypothesis 4: Grandparental involvement is associated with preschoolers’ externalizing problem behaviors through the serial mediating effect of family functioning and emotion regulation (see Fig. 1).

Fig. 1.

Conceptual model of this study

The current study

The quality of caregiving within family systems is significantly associated with children’s development, and the mechanisms underlying this association may be diverse and complex. Although some theoretical perspectives and empirical investigations have examined the role of grandparental caregiving, relatively few studies have directly investigated the relationship between grandparental involvement and preschool children’s externalizing problem behaviors. Given the increasing prevalence of children’s problem behaviors in recent years and the global concern surrounding these issues [61, 62], identifying more effective strategies to prevent and reduce such behaviors has become increasingly urgent. As primary caregivers for many children, grandparents play a pivotal role in facilitating children’s social adaptation and developmental outcomes [12]. Therefore, the present study explores the mechanisms through which grandparental involvement in caregiving is associated with preschool children’s externalizing problem behaviors, with particular attention to the sequential mediating roles of family functioning and emotion regulation. This research advances the field by providing new empirical evidence and offers practical implications for family caregiving practices.

Methods

Participants and procedure

Considering the influence of grandparent involvement on children’s developmental outcomes [63–65], and in accordance with the sampling rationale described above, this study employed a convenience sampling method to recruit participants, with the sampling scope limited to kindergartens in Guangzhou. During the sampling process, balanced distribution, institutional scale, and willingness to cooperate across districts in Guangzhou were carefully considered. Priority was given to public and inclusive private kindergartens covering all 11 urban districts, and 20 kindergartens were ultimately selected as sampling units to ensure a certain degree of regional representativeness. In total, 20 kindergartens in Guangzhou were included for convenience sampling. A total of 862 children from senior, middle, and junior classes who were co-reared by grandparents and parents were recruited. To maintain scientific rigor, each participant received an incentive of 30 RMB (3.66 EUR) in accordance with the approved project budget standards. Designated class teachers distributed a comprehensive questionnaire to the child’s primary caregivers (either parents or grandparents) to collect information on demographics, family functioning, grandparent involvement, children’s emotion regulation, and behavior problems. The recruitment and data collection procedures were approved by the Ethics Committee of the authors’ university.

After data collection, 60 invalid questionnaires were excluded, yielding a final sample of 802 valid cases of children co-reared by grandparents and parents. The exclusion criteria were as follows: (1) questionnaires in which the child was solely cared for by grandparents or solely by parents; (2) questionnaires without identification of the “primary caregiver”; (3) questionnaires containing three or more missing responses regarding the “primary caregiver”; (4) questionnaires missing both “gender” and “age” information of the child; (5) questionnaires with more than one-third of items unanswered (i.e., blank); (6) among questionnaires with identical response patterns, only one was retained as valid, and the remainder were deemed invalid. For valid questionnaires with scattered missing data (missing rate < 5%), multiple imputation was applied for data completion [66]. This method was chosen because it preserves sample size and reduces bias more effectively than mean imputation or listwise deletion [67]. The final sample comprised 802 participants, of whom 42.3% were parents and 57.7% were grandparents. The mean age of the children was 4.83 years (standard deviation = 1.13), with boys representing 50.2% and girls 49.8% of the sample.

Measures

Grandparent involvement

Grandparent involvement was measured using the Grandparent Involvement Questionnaire developed by the Institute of Social Science Survey in China in 2012 [68], which specifically evaluates grandparent caregiving. Response options ranged from 1 (never) to 5 (almost always). Participants were instructed to complete the questionnaire with reference to the grandparent who played the most substantial role in the child’s intellectual and social activities. Cronbach’s alpha for this scale was 0.887.

Externalizing behavioral tendencies

Children’s externalizing tendencies were assessed using the Children’s Behavior Questionnaire-Short Form (CBQ-SF). Although the CBQ primarily assesses temperamental dimensions, prior research on children’s behaviors in Chinese multigenerational families indicates that temperamental reactivity and irritability significantly influence children’s responsiveness to grandparental involvement. These characteristics help explain why comparable levels of grandparental involvement may result in different degrees of internalizing and externalizing problem behaviors [69, 70]. In this study, three core dimensions were examined: surgency (12 items), negative affectivity (12 items), and effortful control (12 items). These dimensions are regarded as key temperamental indicators of externalizing tendencies and provide a reliable framework for examining how grandparental involvement affects child outcomes through family functioning and emotion regulation. Items were rated on a 5-point scale, and the Cronbach’s alpha for the combined measure was 0.915.

Emotion regulation

The Emotion Regulation Checklist was used to evaluate children’s emotion regulation. Parents rated their children’s emotion regulation on a 24-item Likert scale [71], with response options ranging from 1 (never) to 5 (almost always). Mean scores were computed (Cronbach’s alpha = 0.852), with higher scores indicating stronger emotion regulation abilities.

Family functioning

Family functioning was assessed using the General Functioning (GF) subscale of the McMaster Family Assessment Device (FAD). An adapted 12-item version was employed, in which all items were framed to reflect positive family functioning. To ensure that the scale captured the intergenerational dynamics of three-generation households rather than focusing solely on the parental dyad, item wording was modified. Specifically, the pronoun “We” in the original instrument was replaced with “Our family members” (e.g., “Our family members can express feelings to each other”). Although the original FAD-GF subscale uses “We” to denote the family unit, this modification explicitly included grandparents in three-generation households. This adaptation ensures that respondents consider broader intergenerational dynamics rather than only the parental subsystem, thereby enhancing content validity for the present sample. To empirically assess the structural validity of the adapted scale, confirmatory factor analysis was conducted. The results indicated a good model fit (χ² = 69.71, df = 54, CFI = 0.998, TLI = 0.997, RMSEA = 0.018, SRMR = 0.013), suggesting that the modified items reliably loaded onto the intended general functioning construct. Each item was rated on a 5-point Likert scale ranging from 1 (very inconsistent) to 5 (very consistent). Higher mean scores represent better overall family functioning from the perspective of the primary caregiver. Cronbach’s alpha of the scale in this study was 0.813.

Data analysis

In this study, SPSS 27.0 was employed for data processing and analysis. Initially, descriptive statistics (mean and standard deviation) were calculated, and standardization was applied to the independent variable (grandparent involvement), mediating variables (emotion regulation, family functioning), and dependent variable (externalizing problem behaviors). Subsequently, Pearson correlation analysis was conducted to examine the bivariate associations between externalizing problem behavior and the independent and mediating variables. The correlation coefficient r was used to indicate the strength of these associations, with p < 0.05 considered statistically significant. Next, Model 6 of the PROCESS macro in SPSS was applied to evaluate the mediating effects of emotion regulation and family functioning on the relationship between grandparent involvement and externalizing child problem behaviors [72]. Standardized path coefficients β were reported, and p < 0.05 was regarded as statistically significant. Furthermore, because prior research has demonstrated that the bias-corrected percentile Bootstrap method is more robust than the traditional Sobel method [73], the mediation hypothesis was tested using the bias-corrected Bootstrap approach with 5000 bootstrap samples. Gender, age, caregiver role, and caregiver education level were included as covariates in the model. To evaluate the relative contribution of the mediating pathways, the proportion mediated (PM), reflecting the proportion of the total effect explained by indirect effects, was calculated [74]. Indirect effects were considered statistically significant when the 95% confidence interval (CI) did not include 0.

Results

Common method biases tests

Data in this study were obtained through self-reported questionnaires, which may introduce common method bias and potentially influence the findings. To mitigate this issue, several procedural remedies were implemented, including ensuring anonymity, separating different questionnaires, incorporating reverse-coded items, and emphasizing data confidentiality. In addition, the Harman single-factor test was conducted as a post hoc statistical assessment. The results indicated that 14 factors had eigenvalues greater than 1, with the largest factor accounting for 23.287% of the total variance (< 40%). Therefore, common method bias was not considered to pose a substantial threat to the validity of the study results.

Preliminary analysis

The analysis began with a comprehensive assessment of the measurement model to verify the psychometric properties of the research constructs. As shown in Table 1, all scales demonstrated satisfactory internal consistency, with Cronbach’s alpha values ranging from 0.813 to 0.915 and composite reliability (CR) values between 0.741 and 0.984. Convergent validity was supported by average variance extracted values ranging from 0.602 to 0.955, all exceeding the recommended threshold of 0.50 [75]. Discriminant validity was also established, as the square root of the average variance extracted (AVE) for each construct exceeded its correlations with other variables. After confirming the adequacy of the measurement model, Pearson correlation analysis was performed to examine associations among demographic characteristics and the primary study variables. As reported in Table 1, grandparental involvement was positively associated with emotion regulation (p < 0.01) and family functioning (p < 0.01), and negatively associated with externalizing problem behaviors (p < 0.01). Family functioning showed a positive association with emotion regulation (p < 0.01) and a negative association with externalizing problem behaviors in preschool children (p < 0.01). Moreover, emotion regulation was significantly negatively associated with externalizing problem behaviors (p < 0.01). These preliminary results provide an empirical basis for further testing of the proposed mediation pathways.

Table 1.

Descriptive Statistics, Convergent Validity, Discriminant Validity and Correlation Analysis

Variable	M	SD	AVE	CR	1	2	3	4	5	6	7	8
1.Parents education level	2.22	1.148			-
2.Caregiver role	1.77	0.418			0.033	-
3.Children gender	1.48	0.500			0.024	-0.006	-
4.Children age	4.83	1.130			0.057	-0.031	-0.030	-
5.Grandparent involvement	2.63	1.050	0.741	0.602	0.104*	0.094*	0.041	-0.021	0.887
6.Family functioning	1.65	1.020	0.971	0.917	0.059	-0.041	0.045	-0.054	0.261**	0.813
7.Emotion regulation	1.69	0.920	0.950	0.863	0.040	-0.034	0.075	-0.026	0.394**	0.322**	0.852
8.Externalizing problem behaviors	2.33	1.070	0.984	0.955	-0.010	0.030	-0.030	0.064	-0.323**	-0.412**	-0.297**	0.915

Note: N = 802, ^*p < 0.05, ^**p < 0.01, ^***p < 0.001. The diagonal elements represent Cronbach’s alpha coefficients

AVE Average Variance Extracted, CR Composite Reliability

Multiple mediating model analysis

Using Hayes’ SPSS macro PROCESS (Model 6), all continuous variables were standardized (z-scores) prior to mediation analysis. Although standardization is not mandatory, it facilitates interpretation and comparison of path coefficients across mediators and outcomes [76]. The regression models controlled for gender, age, caregiver role, and parental education level. The results are presented in Table 2.

Table 2.

Regression Analysis of the Relationship Between Grandparental involvement and externalizing problem behaviors

Result Variable	Predictor Variable	R	R 2	F	β	t
Externalizing problem behaviors	Caregiver’ education level	0.335	0.112	59.688**	0.018	0.413
	Caregiver role				0.062	1.472
	Children gender				-0.025	0.550
	Children age				0.057	1.329
	Grandparent involvement				-0.329	-7.726**
Family functioning	Caregiver’ education level	0.281	0.079	8.552***	0.030	0.787
	Caregiver role				-1.647	-1.556
	Children gender				0.142	1.617
	Children age				-0.022	-0.569
	Grandparent involvement				0.257	6.037***
Emotion regulation	Caregiver’ education level	0.456	0.217	22.978***	-0.004	-0.139
	Caregiver role				-0.123	0.088
	Children gender				0.041	0.561
	Children age				-0.034	-1.049
	Grandparent involvement				0.296	8.105***
	Family functioning				0.215	5.791***
Externalizing problem behavior	Caregiver’ education level	0.474	0.224	20.567***	0.212	0.592
	Caregiver role				0.079	0.077
	Children gender				-0.006	-0.079
	Children age				0.038	1.005
	Grandparent involvement				-0.177	-3.930**
	Family functioning				-0.161	-3.640**
	Emotion regulation				-0.337	-6.499**

Note: ^*p < 0.05, ^**p < 0.01, ^***p < 0.001. Coefficients are standardized

Grandparent involvement significantly predicted externalizing problem behaviors in preschool children (β = − 0.329, p < 0.01), supporting Hypothesis 1. It also significantly predicted emotion regulation (β = 0.296, p < 0.01) and family functioning (β = 0.257, p < 0.01). Family functioning significantly predicted emotion regulation (β = 0.215, p < 0.01) and externalizing problem behaviors in preschool children (β = − 0.161, p < 0.01). In addition, emotion regulation significantly predicted externalizing problem behaviors in preschool children (β = − 0.337, p < 0.01). These results indicate that family functioning and emotion regulation serve as significant mediators in the relationship between grandparent involvement and externalizing problem behaviors in preschool children. Furthermore, a significant serial mediating effect was observed, thereby supporting Hypotheses 2–4.

The mediating effects were estimated using 5000 bootstrap samples with a 95% confidence interval. The findings demonstrated that emotion regulation and family functioning mediated the relationship between grandparent involvement and externalizing problem behaviors in preschool children. As shown in Table 3, the 95% confidence intervals for all three indirect pathways excluded 0, indicating significant mediating effects through family functioning (indirect effect = − 0.041, Boot SE = 0.013, 95% CI [–0.068, − 0.019]), emotion regulation (indirect effect = − 0.100, Boot SE = 0.019, 95% CI [–0.140, − 0.065]), and the sequential pathway of family functioning and emotion regulation (indirect effect = − 0.019, Boot SE = 0.005, 95% CI [–0.031, − 0.011]). The total indirect effect of grandparent involvement on externalizing problem behaviors was − 0.161 (Boot SE = 0.023, 95% CI [–0.206, − 0.117]), accounting for 48.35% of the total effect. Thus, Hypotheses 2–4 were supported.

Table 3.

Mediation Effect Analysis

Total indirect effect	Indirect Effects	Boot SE	95%CI	Proportion Mediated
	-0.161	0.023	[-0.206,-0.117]	48.35%
Grandparent involvement → Family functioning → Externalizing problem behaviors	-0.041	0.013	[-0.068,-0.019]	12.59%
Grandparent involvement → Emotion regulation → Externalizing problem behaviors	-0.100	0.019	[-0.140,-0.065]	29.94%
Grandparent involvement → Family functioning → Emotion regulation → Externalizing problem behaviors	-0.019	0.005	[-0.031,-0.011]	5.78%

Note: Data are presented as point estimate (95% confidence interval)

The Proportion Mediated (PM) represents the proportion of the total effect of grandparent involvement on externalizing problem behaviors explained by the mediating pathways

Abbreviations: CI Confidence interval, SE Standard error

Discussion

Grandparent involvement and externalizing problem behaviors

Previous research examining the relationship between grandparent involvement and externalizing behavioral tendencies has produced inconsistent results. For example, Hamilton reported that grandparent involvement effectively reduced such behaviors among children in Black families [77]. However, this finding has not been consistently confirmed in Asian family contexts. Another study identified a significant association between grandparent involvement and reduced externalizing tendencies among preschool children in three-generation Hindu families in Asia [78]. The present study addresses the call by Goh and Kuczynski to further investigate how grandparent involvement in child-rearing influences children’s behavior across diverse cultural and family structures [79], particularly within Chinese three-generation households. Unlike Western nuclear family systems, where grandparental involvement may be perceived as a source of intergenerational conflict, the current findings suggest that grandparents in Chinese three-generation households function as integral supportive members. Grandparents not only provide daily care but also broaden preschool children’s cognitive perspectives through the transmission of traditional cultural values, thereby enhancing children’s understanding of social norms and reducing the likelihood of externalizing behavioral tendencies. This culturally embedded motivation is associated with high levels of warmth and “meticulous care” that characterize Chinese caregiving practices. Furthermore, the findings offer empirical support for the “Cha Xu Ge Ju” theory by demonstrating how the strategic positioning of grandparents within the family network sustains systemic flexibility. Accordingly, the role of grandparents in shaping family dynamics may represent a culturally endorsed mechanism that facilitates children’s social adjustment.

It should be acknowledged that although the hypothesized association was statistically significant, the model accounted for a relatively modest proportion of variance in externalizing behavioral tendencies, with an R² of 0.112. This limited explained variance may be attributable to the conceptual nature of the measurement instrument. Because the CBQ-SF assesses temperamental traits rather than clinically defined behavioral symptoms, it captures behavioral precursors or vulnerability markers rather than the full manifestation of behavioral disorders. This interpretation aligns with existing literature indicating that externalizing behaviors are multifaceted and shaped by multiple interacting determinants [80, 81]. While temperament-based instruments such as the CBQ are sensitive to early developmental indicators of behavioral vulnerability [82, 83], additional unmeasured factors, including parental discipline practices, peer interactions, and neighborhood context, likely contribute to these outcomes. Therefore, the present findings should be understood as reflecting early developmental pathways of behavioral vulnerability within the family system rather than clinical diagnoses of behavioral disorders.

The mediating role of family functioning

Grounded in family systems theory, this study confirmed that grandparent involvement influences preschool children’s externalizing problem behaviors through family functioning. This finding supports Marsh et al. [36], who demonstrated that family functioning is significantly associated with early externalizing problem behaviors in children. In well-functioning families, close relationships among members are significantly associated with fewer learning, anxiety, and psychosomatic problems in children. Grandparents voluntarily assume part of the parental caregiving responsibilities, including household chores, cooking, cleaning, and school drop-offs and pick-ups, thereby facilitating stable household functioning. Such support enables mothers to better manage the balance between family and work roles, which is associated with lower levels of family dysfunction [79]. Moreover, effective family systems contribute to the development of secure attachment relationships between preschool children and their caregivers [38], and positive interactions among co-caregivers, serving as social models, are associated with enhanced interpersonal competence and cognitive flexibility, which in turn are linked to reduced externalizing problem behaviors [84].

The mediating role of emotion regulation

This study confirmed that grandparental involvement is negatively associated with preschoolers’ externalizing problem behaviors through the mediating role of emotion regulation. This result is consistent with emotional socialization theory, which posits that children’s emotional competence develops primarily through interactions within the family ecological context [85]. While prior research has largely emphasized how maternal emotion coaching promotes adolescents’ regulation abilities or how parental emotional dysregulation is associated with elevated externalizing problem behaviors, the present findings extend this perspective to non-parental caregivers in three-generation households [86, 87].

Within the Chinese cultural context, grandparents frequently provide attentive care characterized by warmth and emotional support [88]. From a family systems perspective, such a nurturing environment functions as an emotional buffer that appears to promote positive internal working models, thereby strengthening children’s capacity to monitor and regulate emotional experiences. By responding sensitively to children’s distress, grandparents act as secondary socialization agents who model adaptive emotional responses [89]. This developmental process indicates that emotion regulation represents not only an individual characteristic but also a proximal mechanism through which the systemic advantages of grandparental support are internalized, ultimately associated with lower levels of impulsive or aggressive behaviors. Accordingly, promoting children’s emotional competencies constitutes a key protective resource in reducing behavioral risks within contemporary family structures.

The chain-mediating role of family functioning and emotion regulation

The findings suggest that the negative association between grandparental involvement and preschoolers’ externalizing problem behaviors is sequentially mediated by family functioning and children’s emotion regulation. This result is consistent with the cascading effects model in developmental psychopathology [53], which proposes that individual psychological development is embedded within the quality of the broader ecological system [90]. When interpreting these findings in relation to recent literature, cultural distinctions become evident. Several studies conducted in Western contexts have indicated that high levels of grandparental involvement or multigenerational living may be associated with boundary ambiguity and intergenerational conflict, potentially undermining family functioning [91, 92]. In contrast, the present results align with studies in Asian family systems, which consistently report that normative intergenerational coparenting serves as a systemic buffer positively associated with family cohesion and adaptability [12, 18, 34].

Building upon this proximal family context, recent evidence emphasizes that a cohesive family environment is a foundational condition for effective emotional socialization. A stable family system expands the child’s emotion regulation network by providing multiple opportunities for coregulation and observational learning [22, 25]. Consequently, these internalized regulatory capacities operate as proximal mechanisms that are negatively associated with the emergence of externalizing behaviors [21, 23, 26, 27]. The proposed serial mediation model offers an important theoretical refinement of family systems theory [54]. Whereas earlier research often examined grandparenting, family climate, and child behavior in isolated dyadic relationships, the present study delineates a macro-to-micro spillover mechanism specific to three-generation households. It demonstrates that the structural contribution of the extended family (distal factor) does not directly determine child behavior; rather, it operates through the mesosystem of family functioning, which provides the essential foundation for the development of children’s internal emotional competence. This framework advances theoretical understanding of how systemic flexibility is sustained in contemporary intergenerational caregiving contexts.

Practical implications

The findings of this study have important implications for family caregiving practices. First, this research provides valuable insights for advancing intervention strategies targeting preschoolers’ externalizing problem behaviors. The results indicate that grandparental involvement is negatively associated with externalizing problem behaviors, underscoring the constructive role of grandparents within the family caregiving system [93]. Therefore, interventions addressing preschoolers’ externalizing problems may benefit from actively promoting positive grandparental involvement rather than focusing exclusively on parental roles. In family-based interventions, structured workshops can be developed to guide three-generation families in establishing collaborative caregiving models. Such workshops may include sessions on intergenerational communication, shared caregiving responsibilities, and conflict resolution strategies, aiming to support grandparents, parents, and children in forming a harmonious caregiving alliance consistent with the characteristics of Chinese three-generation households.

Second, the serial mediating model identified in this study suggests a more targeted approach for reducing preschoolers’ externalizing problem behaviors by simultaneously strengthening family functioning and enhancing children’s emotion regulation abilities. This perspective is particularly relevant for the design and refinement of early childhood programs. For example, communities and kindergartens may implement structured emotion-focused programs involving three-generation families, such as interactive picture book reading, emotion-based role-play activities, and joint artistic creation among grandparents, parents, and children. Participation of multiple generations in these structured activities may not only enhance emotional bonds and optimize family functioning [94] but also facilitate children’s development of skills in emotion recognition, expression, and regulation within natural interactive contexts [24, 95].

Third, early childhood education institutions may incorporate emotion regulation training into routine teaching and parent-child activities while encouraging grandparents to attend parent-teacher meetings and family education seminars. This approach enables grandparents to better understand the significance of emotion regulation for preschoolers’ development and acquire practical guidance strategies, thereby extending the benefits of early childhood programs into the family environment. By cultivating preschoolers’ emotion regulation as a protective resource [24], such interventions may enhance their capacity to cope with developmental stressors and reduce the association between these stressors and the risk of externalizing problem behaviors. Furthermore, relevant authorities may promote the dissemination of family-based intervention models and early childhood programs within communities and kindergartens, particularly in regions with a high prevalence of three-generation families, in order to maximize the practical value of the present findings.

Limitations

Despite its contributions, several limitations of the present study should be acknowledged. First, regarding measurement, externalizing problem behaviors were assessed using the Children’s Behavior Questionnaire (CBQ), which focuses on dispositional temperament traits rather than observable behavioral or clinical symptom profiles [96]. Second, the reliance on self-reported data introduces potential vulnerability to common method bias; although Harman’s single-factor test was performed, this technique has limited sensitivity in fully excluding such bias [97, 98]. Third, the cross-sectional correlational design restricts interpretation, as causal relationships among variables cannot be established. Fourth, potential confounding variables were not comprehensively controlled, particularly in relation to the multidimensional nature of socioeconomic status; although parental education was included, additional indicators such as household income and occupational status may have influenced the observed relationships [99, 100]. Finally, the generalizability of the findings may be constrained by cultural specificity, as the study focused on Chinese three-generation families within a context shaped by norms of filial piety, which may limit applicability to other cultural settings.

Future research directions

To address the limitations of the current study and further advance this field, future research should refine assessment strategies, strengthen causal inference, and expand examination of contextual influences. Specifically, subsequent studies may employ clinical instruments such as the Child Behavior Checklist (CBCL), the Eyberg Child Behavior Inventory (ECBI), or the Strengths and Difficulties Questionnaire (SDQ) to further validate the present findings [101–104]. In addition, to enhance objectivity and reduce common method bias, future research should adopt multi-informant designs that integrate reports from parents, teachers, and grandparents, supplemented by direct behavioral observations [105, 106]. Moreover, longitudinal or experimental designs should be implemented to clarify causal pathways and examine developmental changes over time. Incorporating multiple indicators of socioeconomic status as covariates would improve model precision, and testing the proposed model across diverse cultural contexts is recommended to evaluate cross-cultural generalizability [107].

Conclusion

This study provides a comprehensive examination of how family functioning and children’s emotion regulation mediate the relationship between grandparents’ involvement and preschool children’s externalizing problem behaviors. The findings emphasize the distinctive value of grandparents within the family system. Specifically, the beneficial association of grandparents’ involvement with children’s behavior appears to operate through optimizing the broader family environment and strengthening children’s internal emotional competencies. The results suggest that when grandparents participate actively and positively, they may enhance family communication and emotional cohesion. This, in turn, may promote children’s emotion regulation competence, which is associated with reduced externalizing behaviors. Accordingly, these findings highlight the importance of moving beyond the traditional parent-child dyad and incorporating grandparents into family-based parenting interventions. Efforts to improve family functioning and guide grandparents’ effective participation should be prioritized to establish a supportive developmental ecosystem for children. To the best of our knowledge, this study is among the first to empirically examine this serial mediation mechanism within the framework of family systems theory, addressing a gap concerning intergenerational caregiving and offering practical insights for collaborative education between families and kindergartens.

Authors’ contributions

YX and XC: developing the theoretical framework, data analysis, organisation, and overall writing of the paper; CL and YN: editing and organisation of the paper as well as the overall design; YW and TG: drafting the work and revising critically.

Funding

This research was funded by the 2025 Guangdong Philosophy and Social Science Planning Project (GD25CJY13).

Data availability

The data used to support the findings of this study are available from the corresponding author upon request.

Declarations

Ethics approval and consent to particpate

This study complies with the Declaration of Helsinki. Ethical approval was obtained from the research ethics committee of the School of Educational Science at Guangdong Polytechnic Normal University (JKYLLSM20230057) before data collection.

Informed consent was obtained from the legal guardians of all child participants for study participation. No identifying information or images have been included in this manuscript. All personal identifiers have been removed from the text, figures, tables, and images to ensure the anonymity of the participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.