Beyond one's own child: progressive parenting beliefs and the cross-household socialization of children's prosocial behavior

Children's social development unfolds within multiple interconnected contexts, and parenting beliefs play a central role in shaping children's social behavior, yet little is known about whether such influences extend beyond the family to affect children in peer networks. This study examined whether the progressive parenting beliefs of a best friend's mother, characterized by autonomy support and reasoning, were associated with the study child's prosocial behavior via the best friend's prosocial behavior. Using data from the NICHD Study of Early Child Care and Youth Development (N = 836), we tested a mediation model and examined whether the indirect association varied by the study child's individual and contextual characteristics. Results revealed a significant indirect association between best friends' mothers' progressive parenting beliefs and the study child's prosocial behavior, mediated by the best friend's prosocial behavior, with no significant direct effect. Moderated mediation analyses further showed that this indirect association was most evident among children from lower-SES backgrounds, with the conditional indirect effect also present at average SES but not at higher SES. These findings suggest that the influence of progressive parenting beliefs may extend beyond the family household through close peer relationships, underscoring the importance of peer contexts as a source of adaptive socialization in resource-limited environments. The results provide evidence for the socio-ecological nature of prosocial development, showing how progressive parenting beliefs can ripple across interconnected family and peer networks. Interventions to promote progressive parenting beliefs may help increase prosocial behavior not only in targeted children but also in their close friends, particularly among children from socioeconomically disadvantaged contexts.

Children's social development unfolds within multiple interconnected contexts, including families, peer relationships, and broader social environments. Prosocial behaviors such as helping, sharing, and cooperating play a central role in children's social

and moral development. Children who engage in prosocial behaviors tend to experience more positive peer relationships, greater social competence, and stronger adjustment across school and community contexts (Crone and Achterberg, 2022; Padilla-Walker et al., 2018; van Meegen et al., 2024). For these reasons, promoting prosocial behavior has become a central focus of research on positive youth development (PYD), which emphasizes the alignment of contextual resources with youth strengths over time (Carlo and Padilla-Walker, 2020; Meulenbeek et al., 2026; Padilla-Walker et al., 2018). Prosocial behavior does not emerge in isolation. It develops within broader social ecologies, where parents, peers, and other adults together shape how children come to understand others' needs, internalize moral norms, and act with care in everyday life (Bronfenbrenner, 1979). How these interconnected relationships contribute to prosocial development remains a central question for developmental science.

Parenting beliefs and children's prosocial development

A primary context for children's prosocial development is the family environment. Families offer the earliest and most consistent setting in which children learn social norms, moral expectations, and ways of relating to others. Through everyday interactions, parents foster children's understanding of others' needs, model social behavior, and guide children's responses to social situations. Within this broader family context, parenting beliefs are a key mechanism through which children's social behavior develops. Beliefs about childrearing reflect assumptions about autonomy, discipline, and social responsibility, and they shape how parents interpret and respond to their children's behavior (Distefano and Meuwissen, 2022; Grusec and Davidov, 2010). Progressive parenting beliefs typically emphasize empathy, open communication, autonomy support, and perspective-taking, whereas more traditional beliefs place greater emphasis on obedience, conformity, and respect for authority. These belief systems shape everyday parenting practices that influence children's moral orientations and social behaviors.

Research consistently links warm, autonomy-supportive parenting to higher levels of prosocial behavior, including helping, sharing, and perspective-taking, from childhood through adolescence (Carlo et al., 2007; Padilla-Walker et al., 2016; van Meegen et al., 2024; Wong et al., 2021). When parents encourage children to consider others' perspectives and guide behavior through reasoning, children are more likely to internalize prosocial values and enact them across contexts. At the same time, this literature has focused almost exclusively on the influence of children's own parents, leaving open the question of whether parenting beliefs may extend beyond the boundaries of the household.

Peer influence and best friend relationships

Peer relationships are another major context for socialization in childhood. Through processes such as modeling, reinforcement, and mutual influence, children's behaviors become increasingly aligned with those of their peers over time (Choukas-Bradley et al., 2015; Dishion et al., 1991; Dishion and Tipsord, 2011). Within peer networks, relationships vary in closeness and influence. While broader peer groups and classmates are important, close friendships, characterized by trust, intimacy, and frequent interaction, often serve as especially powerful contexts for social learning. Best friends spend substantial time together, share emotionally meaningful experiences, and exert strong mutual influence on one another's attitudes and behaviors (Berndt, 2002; Kindermann, 2007). Over time, they often grow increasingly similar in behaviors and values through socialization and mutual influence (Dishion and Tipsord, 2011; Kindermann, 2007). In a year-long study of adolescents and their reciprocated friends, a friend's prosocial behavior predicted the adolescent's own prosocial behavior over time, a link explained by the prosocial goals the adolescent adopted (Barry and Wentzel, 2006). Such closeness and mutual influence make best friendships especially potent contexts in which prosocial behaviors such as cooperation, empathy, and helping are reinforced and transmitted. Accordingly, best friendships provide a particularly meaningful context for examining how developmental processes may extend across social contexts.

Cross-household socialization

Although parenting and peer influences are both well-documented, they are often examined as distinct domains. From a socio-ecological perspective, children develop within interconnected systems in which family and peer contexts are dynamically linked (Bronfenbrenner and Morris, 2006; Meulenbeek et al., 2026). In this view, close friendships can function as bridges between family environments, forming a mesosystem through which experiences, norms, and behaviors from one household may influence development in another. Best friends frequently spend time in each other's homes and share daily experiences. The adults tied to these friendships, particularly best friends' mothers, may therefore occupy a distinctive and influential position in children's social ecologies. These adults may act as salient “other parents” whose beliefs and practices children regularly observe, experience, and potentially internalize beyond their own families.

Emerging research suggests that parenting influences may extend beyond the parent-child dyad to shape broader peer environments. Parenting practices are associated with children's peer affiliations and with the norms operating within peer contexts (Bagwell and Bukowski, 2018; Vaughan et al., 2022), and parents help shape the social environments in which children interact (Brown and Bakken, 2011; Dodge et al., 2006). To date, this work has focused largely on risk-related outcomes such as delinquency and substance use, and it has often treated peers' parents as a diffuse influence rather than tracing specific relational pathways. Relatively little research has examined whether parenting beliefs in one family contribute to positive developmental outcomes in other children through specific relational pathways such as best friendships.

Socioeconomic status (SES) may further shape these cross-household processes. Children growing up in lower-SES environments often face greater structural constraints and may rely more heavily on relationships outside the immediate family for support and guidance. Supportive peer relationships and prosocial behaviors can function as protective factors that promote positive adjustment in the face of adversity (Eisenberg et al., 2015; Meulenbeek et al., 2026). This protective value appears to be greater in disadvantaged contexts. The benefits of prosocial behavior for children's adjustment are more pronounced among those living in lower-SES neighborhoods or attending low-performing schools (Armstrong-Carter et al., 2021; Flouri and Sarmadi, 2016). In such settings, exposure to peers who model prosocial behavior, potentially shaped by their family environments, may carry amplified developmental significance. In higher-SES contexts, by contrast, multiple sources of support may reduce the relative influence of any single peer pathway. These considerations highlight the importance of examining how broader contextual conditions shape the extent to which parenting beliefs reach beyond the household through peer relationships.

The present study

The present study builds on these perspectives by examining whether parenting beliefs extend beyond the family to influence another child's development through best friend relationships. Specifically, it tests whether a best friend's mother's progressive parenting beliefs are associated with the study child's prosocial behavior, both directly and indirectly through the best friend's own prosocial behavior (see Figure 1). The study also explores whether these associations vary across individual and contextual conditions, including children's sex, race/ethnicity, SES, and parenting beliefs within the study child's own family.

Research questions

The present study addressed two primary research questions.

Are the best friend's mother's progressive parenting beliefs associated with the study child's prosocial behavior directly and indirectly through the best friend's prosocial behavior?

Does this indirect association vary across children's sex, race/ethnicity, SES, and the study child's own mother's parenting beliefs?

Drawing on research in parenting, peer influence, and socio-ecological perspectives, the present study makes several contributions. First, it advances understanding of cross-household socialization by identifying a specific pathway through which parenting beliefs in one family may reach children beyond the immediate household. Second, it highlights best friendships as a key relational mechanism through which developmental processes may be transmitted across social networks. Third, it situates these processes within a broader socio-ecological framework by examining how their strength varies across socioeconomic contexts. Taken together, these contributions extend our understanding of child social development by showing how family and peer contexts interact to influence children's prosocial behavior. In doing so, the study seeks to illuminate how contextual belief systems and youth strengths become aligned within children's social ecologies, contributing to a more nuanced understanding of positive youth development.

Participants

The present study utilized data from the National Institute of Child Health and Human Development (NICHD) Study of Early Child Care and Youth Development (SECCYD), a longitudinal study examining child development across multiple contexts, including families, schools, and peer relationships (NICHD Early Child Care Research Network, 2001). The original cohort consisted of 1,364 infants recruited from hospitals in 10 locations across the United States. The present study represents a secondary analysis of these longitudinal data.

The analytic sample included 836 participants with observed Grade 6 target child prosocial behavior. This outcome-observed sample was used as the primary analytic sample to maximize available data for the focal outcome while retaining cases with partially missing predictor or mediator variables.

The sample was approximately evenly distributed by sex (50.4% female). The majority of participants identified as White (82.3%), followed by Black/African American (11.1%), Asian/Pacific Islander (1.3%), American Indian (0.4%), and other racial/ethnic groups (4.9%). Hispanic ethnicity was assessed separately from race; participants identifying as Hispanic were included in the non-White category for analytic purposes. Because several racial/ethnic groups were too small to model individually, we created a composite variable contrasting White with non-White participants to preserve statistical power and model stability. After Hispanic-identifying participants were grouped into the non-White category, this composite yielded 78.1% White and 21.9% non-White participants in the final analytic sample.

SES, indexed by the income-to-needs ratio at Grade 6, had a mean of 4.48 (SD = 3.80), indicating that, on average, families were above the poverty threshold, with substantial variability across participants.

Procedure

At the Grade 6 assessment, study children participated in a laboratory visit that included individual assessments and structured interaction tasks with a nominated best friend. To identify the nominated best friend, study mothers were contacted and asked to consult with their child to nominate up to three close friends (one primary and two backup candidates) to accompany the child to the lab. Each candidate was evaluated using standardized criteria (age within ±2 years [required], same sex, acquaintance of at least 6 weeks, and regular contact at least once per week), and the highest-ranked eligible friend was selected, with backup options retained if needed (NICHD, 2004a).

Data were collected from multiple sources, including the study child, the nominated best friend, and both children's mothers via questionnaire measures. Prosocial behavior for both the study child and the best friend was assessed at Grade 6 using teacher reports. Demographic information for the study child (e.g., sex and race/ethnicity) was collected during the 1-month home visit. SES, indexed by the household income-to-needs ratio, was assessed at Grade 6.

To align with data availability and maintain conceptual clarity, variables were drawn from multiple waves of the NICHD Study of Early Child Care and Youth Development. Parenting beliefs for the study child were assessed at Grade 4, the most recent assessment prior to Grade 6, as this construct was not measured at Grades 5 or 6. Parenting beliefs for the best friend were obtained at Grade 6 to align with the timing of the best friend assessment.

Measures

Prosocial behavior

Prosocial behavior toward peers, both among study children and their best friends, was measured using the Perceptions of Children's Peer-Related Behaviors questionnaire. Teachers completed this questionnaire for each student during the data collection period, rating the extent to which each statement reflected the student's behaviors toward their peers in the school context.

This questionnaire contains a nine-item prosocial behavior toward peers scale, which includes five items from the Child Behavior Scale (Ladd and Profilet, 1996) and four additional items added by NICHD researchers. Responses were rated on a three-point Likert scale from 0 (not true) to 2 (often true). Sample items include “[The child] is kind toward peers” and “[The child] offers help or comfort when other children are upset.” Internal reliability was reported as high for this measure (α = 0.91–0.92; NICHD, 2004a).

Parenting beliefs

Parenting beliefs were assessed using the Parental Modernity Scale (Schaefer and Edgerton, 1985), which captures variation in progressive vs. traditional childrearing beliefs. Progressive beliefs reflect autonomy support and recognition of children's perspectives, whereas traditional beliefs emphasize obedience and parental authority.

Parenting beliefs for the study child's mother were assessed at Grade 4, the most recent assessment prior to Grade 6, as this construct was not measured at Grades 5 or 6. Parenting beliefs for the best friend's mother were assessed at Grade 6 to align with the timing of the best friend assessment.

The scale includes 30 items rated on a five-point Likert scale (1 = strongly disagree to 5 = strongly agree). Sample items include “Children have a right to their own point of view” (progressive) and “Children will not do the right thing unless they must” (traditional). Following standard procedures, scores were computed by combining traditional belief items with reverse-coded progressive items. For clarity in the present study, all items were recoded such that higher scores indicate more progressive parenting beliefs. The measure demonstrates high internal reliability (α = 0.88 −0.94; NICHD, 2004b).

Socioeconomic status

SES was indexed using the household income-to-needs ratio at Grade 6. The income-to-needs ratio was calculated by dividing total household pre-tax income by the federal poverty threshold, adjusted for household size, number of children, and year of assessment, with values below 1 indicating income below the poverty line. Higher values indicate greater socioeconomic advantage.

Data analysis plan

All analyses were conducted in R. Descriptive statistics and correlations were computed for the primary study variables using standard R packages, and mediation and moderated mediation models were estimated using the lavaan package for structural equation modeling. Prior to analysis, missing values coded as −999 were recoded as missing. The primary analytic sample included participants with observed Grade 6 study child prosocial behavior. Missing data on predictors, mediator, and covariates were handled using full information maximum likelihood (FIML), which allows cases with partially missing data to contribute to model estimation under the assumption that data are missing at random. Preliminary analyses were conducted to examine whether missingness was systematically associated with observed variables.

To examine the hypothesized interpersonal pathway, a mediation model was tested in which best friend's mother's progressive parenting beliefs predicted the study child's prosocial behavior indirectly through best friend prosocial behavior. The model included both the indirect pathway through best friend prosocial behavior and the direct pathway from best friend's mother's parenting beliefs to the study child's prosocial behavior. The child's mother's parenting beliefs, SES, sex, and race/ethnicity were included as covariates and were specified as predictors of the study child's prosocial behavior. Sex was coded as 0 = male and 1 = female, and race/ethnicity was coded as 0 = non-White and 1 = White. These covariates reflect characteristics of the study child and their family context and were included to account for individual- and family-level influences on prosocial behavior.

Further, to examine whether the indirect pathway was conditional on child and family contexts, moderated mediation analyses were conducted following the conditional indirect effects framework described by (Preacher et al. 2007). Each moderator was tested in a separate model, including SES, sex, race/ethnicity, and the study child's mother's parenting beliefs. Moderation was specified on the path from best friend prosocial behavior to the study child's prosocial behavior, allowing the indirect effect of best friend's mother's parenting beliefs on the study child's prosocial behavior through best friend prosocial behavior to vary across levels of each moderator. This approach allowed us to examine whether the strength of the overall indirect pathway differed across contexts, rather than focusing only on moderation of individual paths within the model. Continuous predictors and moderators, including SES and the study child's mother's parenting beliefs, were standardized prior to analysis to facilitate model estimation and comparability of coefficients. For continuous moderators, conditional indirect effects were estimated at low (−1 SD), mean, and high (+1 SD) levels. For categorical moderators, conditional indirect effects were estimated separately by group.

Indirect effects and conditional indirect effects were evaluated using bootstrapped confidence intervals based on 1,000 resamples. Additionally, sensitivity analyses were conducted to examine the robustness of the primary findings.

Descriptive statistics

Descriptive statistics and correlations among study variables are presented in Table 1. Study child prosocial behavior was positively associated with best friend prosocial behavior (r = 0.25, p < 0.001), best friend's mother's parenting beliefs (r = 0.14, p < 0.001), and the child's mother's parenting beliefs (r = 0.19, p < 0.001). Best friend prosocial behavior was also positively associated with best friend's mother's parenting beliefs (r = 0.25, p < 0.001).

SES was positively associated with all study variables (rs = 0.15–0.37, ps < 0.001). The child's mother's parenting beliefs were moderately associated with best friend's mother's parenting beliefs (r = 0.34, p < 0.001) and SES (r = 0.37, p < 0.001). All correlations were below 0.40, suggesting that multicollinearity was unlikely to pose a concern.

Missing data analysis

A logistic regression analysis examined whether missingness was systematically associated with study variables. Complete-case status (1 = complete, 0 = incomplete) was not significantly predicted by study child prosocial behavior, parenting beliefs, SES, sex, or race/ethnicity (ps > 0.43). These findings indicate that missingness was not systematically related to observed variables and support the use of full information maximum likelihood (FIML) estimation.

Mediation analysis

The mediation model is depicted in Figure 2, and full path estimates are presented in Table 2. This model tested whether best friend prosocial behavior mediated the association between best friend's mother's parenting beliefs and study child prosocial behavior, controlling for the child's mother's parenting beliefs, SES, sex, and race/ethnicity.

Best friend's mother's progressive parenting beliefs were positively associated with best friend prosocial behavior (b = 0.099, SE = 0.016, p < 0.001), which in turn predicted study child prosocial behavior (b = 0.184, SE = 0.052, p < 0.001). The direct effect was not significant (b = 0.007, SE = 0.018, p = 0.681).

The indirect effect was significant [b = 0.018, SE = 0.006, 95% CI (0.007, 0.032), p = 0.002], indicating a mediation pathway. The total effect was not significant (b = 0.025, SE = 0.017, p = 0.143), consistent with an indirect-only mediation pattern.

Among covariates, the child's mother's parenting beliefs (b = 0.049, p = 0.004), SES (b = 0.034, p = 0.032), and sex (b = 0.228, p < 0.001) were significant predictors, whereas race/ethnicity was not (b = 0.046, p = 0.250). The model accounted for 6.5% of the variance in best friend prosocial behavior and 13.3% of the variance in study child prosocial behavior.

Moderated mediation analysis

The results of conditional indirect effect analyses across moderators are summarized in Table 3. For SES, the conditional indirect effect was significant at low (−1 SD) and mean levels of SES, but not at high levels (+1 SD). The estimated indirect effects decreased from low to high SES. Thus, the indirect pathway was most evident among children from lower-SES backgrounds, with the conditional indirect effect also present at average SES but not at higher SES.

In contrast, the indirect effect remained significant and similar in magnitude across low, mean, and high levels of the child's mother's parenting beliefs, suggesting little evidence that the indirect pathway varied meaningfully as a function of the child's mother's parenting beliefs. For sex, the conditional indirect effect was significant among both males and females, though smaller in magnitude among females, suggesting limited variation by sex. Across race/ethnicity, the conditional indirect effect reached significance among White children but not among non-White children, although estimated indirect effects were comparable across groups.

Overall, these findings suggest that the indirect association was generally robust across individual characteristics, with the clearest contextual pattern observed for SES.

Sensitivity analysis

Findings from the complete-case sample (N = 504) were consistent with the primary analyses. The indirect effect of best friend's mother's parenting beliefs on the study child's prosocial behavior through best friend prosocial behavior remained significant [b = 0.014, 95% CI (0.004, 0.024)]. A similar pattern emerged for SES conditional indirect effects, with significant indirect effects at low and mean levels of SES but not at high levels. These results indicate that the findings were robust to an alternative approach to handling missing data.

As an additional robustness check, maternal marital status was included as an indicator of family structure for both the study child and best friend. Marital status was coded as 1 = married and living with spouse and 0 = all other nonmissing statuses, including cohabiting, separated/divorced, widowed, or other. The study child's mother's marital status was specified as a predictor of the study child's prosocial behavior, and best friend's mother's marital status was specified as a predictor of best friend prosocial behavior. Maternal marital status was positively associated with prosocial behavior for both the study child and best friend. Importantly, the indirect effect remained significant and substantively unchanged [b = 0.015, 95% CI (0.006, 0.025), p = 0.002], indicating that the observed cross-family pathway was not explained by differences in family structure as indexed by maternal marital status.

The present study examined whether parenting beliefs extend beyond the family to influence another child's development through peer relationships. Results supported an indirect pathway linking a best friend's mother's parenting beliefs to children's prosocial behavior through the friend's own prosocial behavior, while the direct association was not significant. Specifically, children whose best friends had mothers who endorsed more progressive parenting beliefs tended to have friends who displayed higher levels of prosocial behavior, which in turn predicted greater prosocial behavior in the study child. These findings suggest that parenting processes are not confined to the immediate parent–child dyad but may operate across interconnected social networks, with peers serving as proximal conduits of influence (Bronfenbrenner and Morris, 2006; Meulenbeek et al., 2026).

An important feature of the present study is its focus on best friend dyads rather than peers more broadly. Best friendships represent particularly close and emotionally salient relationships, characterized by greater interaction frequency, intimacy, and mutual influence compared to general peer relationships (Berndt, 2002; Kindermann, 2007). As such, best friends may serve as especially potent settings for socialization, providing a meaningful context in which to examine how family influences extend beyond the household.

In addition, the magnitude of the indirect association differed by SES, with the pathway being more pronounced among children from lower-SES backgrounds. The results provide empirical support for a cross-household socialization model and extend prior research by demonstrating how positive developmental processes may spread through close peer relationships.

One alternative explanation for this pattern is that children and their families are already similar due to selection processes. The moderate correlation between the best friend's mother's and the study child's mother's parenting beliefs (r = 0.34, r 2 ≈ 0.12) suggests some similarity across families, consistent with homophily, the tendency for individuals to associate with others who are similar in values and backgrounds (McPherson et al., 2001). However, because the two variables shared only about 12% of their variance, they largely reflect independent family influences rather than redundant constructs. Moreover, the indirect association between the best friend's mother's parenting beliefs and the study child's prosocial behavior remained significant even after controlling for the child's mother's parenting beliefs, strengthening the interpretation that cross-family pathways may contribute to the development of children's prosocial behavior.

Cross-household socialization and the diffusion of prosocial behavior

The present findings can be understood in relation to several complementary lines of research, suggesting a more integrated view of children's social development (Bronfenbrenner and Morris, 2006). Prior work on collective efficacy has emphasized that children are influenced not only by their own parents but also by networks of adults who shape the broader social environment. For example, (Sampson et al. 1997) found that communities characterized by stronger informal social control are associated with lower levels of problem behavior. Related work further highlights the role of neighborhood social organization and adult networks in shaping youth development (Sharkey and Faber, 2014). The current findings extend this perspective by suggesting that such influence may also operate at a more proximal level within children's immediate social networks, where the parenting beliefs of a specific peer's parent are associated with children's behavior.

These results also align with research on peer socialization demonstrating that children's behaviors, and prosocial behaviors in particular, are shaped through close friendships via processes of modeling, reinforcement, and social learning (Dishion et al., 1991; van Hoorn et al., 2016). Developmental research has consistently shown that children become more similar to their friends over time in a range of behaviors and attitudes (Bagwell and Bukowski, 2018; Kindermann, 2007). In particular, best friend relationships represent a particularly salient context for socialization, as they are characterized by greater intimacy, frequency of interaction, and behavioral similarity compared to broader peer networks (Berndt, 2002; Güroglu, 2022). While prior studies have largely focused on the spread of problem behaviors, such as delinquency and substance use, emerging research suggests that prosocial behaviors may also diffuse through peer networks (Molano et al., 2013; Van Zalk et al., 2010).

Despite this growing body of work, relatively little research has examined the family environments that shape those peer behaviors. The present findings address this gap by suggesting that parenting beliefs may indirectly influence not only one's own child but also the child's friends. Specifically, progressive parenting beliefs among a child's best friend's mother were associated with higher prosocial behavior in that friend, which in turn predicted the study child's prosocial behavior. This pattern suggests that family socialization processes may spill over across peer networks, pointing to an upstream source of peer influence in which parenting beliefs shape peer behavior and, in turn, influence children beyond the immediate family.

Furthermore, these findings complement research demonstrating that parenting practices shape not only children's behavior but also their peer environments (Brown and Bakken, 2011; Dishion et al., 1991). Prior work has shown that parents influence children's peer affiliations through both direct and indirect processes, including monitoring, guidance, and value socialization (Brown and Bakken, 2011). The present study extends this work by showing that parenting influences may reach beyond one's own child, affecting other children through best friend relationships. From a social learning perspective (Bandura, 1977), children acquire behaviors by observing and interacting with peers who serve as salient behavioral models. When a child's best friend displays prosocial behavior that may have been fostered within their family environment, those behaviors may be transmitted through everyday interactions, thereby extending the influence of family socialization beyond the household.

These findings are also consistent with ecological perspectives on development, which stress the interconnected nature of children's social environments. Developmental systems perspectives similarly highlight the dynamic interplay between individuals and their multiple contexts over time (Bronfenbrenner and Morris, 2006; Meulenbeek et al., 2026). Specifically, the results illustrate how experiences in one context, such as the family, may be linked to and influence development in another context, such as peer relationships.

Extending this perspective, the present findings also resonate with the idea that children's development occurs within interconnected social communities. The widely recognized notion that “it takes a village to raise a child” reflects the understanding that children are shaped not only by their own parents but also by the network of relationships in which they are embedded. Developmental theories similarly emphasize that children grow within overlapping social contexts, including families, peer groups, schools, and communities (Bronfenbrenner and Morris, 2006; Meulenbeek et al., 2026).

The present study provides empirical support for this idea by showing that parenting beliefs in one family may influence not only that family's child but also other children through best friend relationships. This pattern is consistent with research showing that parenting practices and beliefs play a central role in shaping children's prosocial development and moral behavior (Carlo and Padilla-Walker, 2020; van Meegen et al., 2024), as well as research indicating that prosocial behaviors can spread through peer interactions (van Hoorn et al., 2016; Van Zalk et al., 2010). From this perspective, parenting beliefs may help constitute a broader social environment that supports prosocial development, with effects that ripple outward through children's social networks.

At the same time, the present findings raise important questions about the conditions under which cross-household socialization processes may be most evident. The most pronounced contextual pattern emerged for SES, which is discussed in greater detail below. For other family and child characteristics, such as the child's mother's parenting beliefs, sex, and race/ethnicity, the conditional indirect effects showed more limited evidence of meaningful variation across levels or groups.

For race/ethnicity, the conditional indirect effect reached significance among White children but not among non-White children. However, the estimated indirect effects were comparable across groups. This pattern suggests that the non-significant finding among non-White children may reflect reduced statistical power due to the smaller sample size rather than a substantive difference in the underlying process. Thus, the cross-household socialization pathway for prosocial behavior may operate similarly across race/ethnicity groups, although further research with more diverse and balanced samples is needed to draw more definitive conclusions.

Socioeconomic context and compensatory processes

The SES pattern provides further insight into the conditions under which cross-household socialization processes may be most evident. The conditional indirect effect was significant at low and mean levels of SES but not at high SES, and the estimated indirect effects decreased from low to high SES. This pattern suggests that peer-related family influences may play a more prominent role in contexts where socioeconomic resources are more limited. This interpretation is consistent with ecological perspectives, which emphasize that in contexts of limited resources, children may rely more heavily on external social networks, including peers and other adults, for guidance and support (Bronfenbrenner and Morris, 2006). In such settings, prosocial behavior may function as a protective factor that promotes adaptive development despite environmental constraints (Armstrong-Carter et al., 2021; Flouri and Sarmadi, 2016). Exposure to prosocial peers may further support children's own prosocial development through peer socialization and social learning processes (Busching and Krahé, 2020).

This interpretation aligns with evidence that the protective benefits of prosocial behavior are themselves amplified in disadvantaged contexts (Armstrong-Carter et al., 2021; Flouri and Sarmadi, 2016). The present findings extend that pattern from a child's own prosocial behavior to the prosocial behavior of a close friend, suggesting that the heightened developmental value of prosociality in low-SES settings may operate across household boundaries as well as within the individual child. As a result, exposure to peers who model prosocial behavior, shaped in part by supportive and autonomy-oriented parenting beliefs, may be especially salient for children's own prosocial development in lower-SES contexts. These findings extend socio-ecological models by showing that the influence of parenting beliefs may cross household boundaries, and that such cross-context processes may be particularly consequential for children growing up in lower-SES environments.

Building on this perspective, prior work on collective efficacy and the social capital perspective further suggests that shared norms and informal social control within communities may play a particularly important role in disadvantaged contexts, where such processes can help compensate for structural constraints (Coleman, 1988; Sampson et al., 1997; Small, 2009). Within this framework, a friend's parent may serve as an additional source of socialization, reinforcing or supplementing the developmental inputs available within the child's own family. Exposure to peers from families that endorse supportive and socially oriented parenting beliefs may therefore promote prosocial development.

These findings are most directly aligned with the concept of compensatory socialization, in which experiences outside the family may partly offset differences in family resources or opportunities (Crosnoe, 2009; Lareau, 2011). Peer contexts, including best friend relationships, may provide opportunities for social learning that supplement or offset variation in family environments.

Extending this line of reasoning, developmental niche theory further suggests that children's development is shaped by culturally and economically structured environments that influence everyday experiences and opportunities for social learning (Super and Harkness, 1986). In this sense, children growing up in different SES contexts may rely differently on extra-familial socialization environments, including peer relationships.

These perspectives suggest that cross-household socialization may act as a compensatory process in disadvantaged environments, with potential relevance for narrowing disparities in developmental outcomes. Although the indirect effect was modest in magnitude, it was statistically robust and theoretically meaningful, and the SES pattern represents a particularly novel contribution.

Implications for theory and practice

The present study makes a novel contribution by identifying a cross-household pathway of socialization, demonstrating that parenting beliefs in one family may influence children beyond the immediate household through best friend relationships. It does so by integrating perspectives on collective socialization, peer contagion, and parenting into a more unified framework. Rather than treating these influences as independent, the findings suggest that they operate through interconnected pathways, with parenting effects extending across social networks via peer relationships. By linking parenting processes to peer socialization, the findings extend existing models of development to account for how prosocial behaviors may diffuse across children's social networks. This perspective underscores the importance of mesosystem processes within bioecological models of development (Bronfenbrenner and Morris, 2006), in which interactions across settings play a central role in shaping developmental outcomes. These findings are also consistent with social capital perspectives, which emphasize that resources embedded in social relationships can shape developmental outcomes (Coleman, 1988), as well as research on peer socialization showing that children's behaviors are influenced by their close peers (Dishion and Tipsord, 2011; van Hoorn et al., 2016).

In terms of practical implications, these findings suggest that interventions targeting parenting beliefs and practices may have broader impacts than previously recognized. Programs designed to promote autonomy-supportive and prosocial parenting have been shown to foster positive developmental outcomes (Carlo and Padilla-Walker, 2020; van Meegen et al., 2024), and the present results suggest that their influence may extend beyond the targeted child to their close peers. This raises the possibility that relatively localized efforts could generate ripple effects across peer groups, amplifying their overall impact. Moreover, these processes may be particularly pronounced among children from lower-SES backgrounds, suggesting that cross-household socialization may serve a compensatory role in contexts of limited resources (Sampson et al., 1997; Small, 2009). Together, these findings highlight the potential of drawing on peer networks and family processes as joined pathways for promoting positive developmental outcomes.

Limitations and future directions

The present study used secondary data from a large archival longitudinal dataset, which provided an unusual opportunity to examine cross-household socialization but also constrained the research design and available measures. As a result, the analyses were limited to variables included in the original data collection, which may have restricted the ability to fully capture relevant developmental and family processes. Accordingly, several limitations should be considered when interpreting these findings and designing future studies. First, although the results support the proposed mediation pathway, causal inferences remain limited because the predictor, mediator, and outcome variables were measured concurrently at Grade 6. Longitudinal and experimental research is needed to clarify the directionality of these processes.

Second, the study was limited by the available informants and methods. Parenting beliefs were assessed through maternal reports, which precluded examination of paternal parenting beliefs and other caregiver influences. Prosocial behavior was assessed using teacher reports of children's peer-related behavior in the school context. Although teacher reports provide valuable information about children's behavior with peers in classroom settings and reduce shared method variance with maternal-reported parenting beliefs, they do not capture prosocial behavior across all contexts. Multi-informants and multi-method assessments across school, family, peer, and community contexts would strengthen future work.

Third, although the model included family SES and the study child's own mother's parenting beliefs, the present analyses could not fully account for broader family-context variables that may overlap with or help explain the observed association between parenting beliefs and prosocial behavior across households. These include family structure and composition, birth order, sibling relationships, and other aspects of family functioning. Sensitivity analyses incorporating maternal marital status, an available indicator of family structure, suggested that the indirect pathway was not explained by this factor. Incorporating a broader range of family characteristics would help clarify the specificity and robustness of the cross-household pathway identified here.

Fourth, the sample included disproportionate representation across racial and ethnic groups, with relatively fewer non-White participants, which may have limited statistical power to detect subgroup differences (Cohen, 1988; Maxwell, 2004). Relatedly, the use of a composite race/ethnicity variable (White vs. non-White) may have further obscured meaningful heterogeneity across groups. More broadly, limited representation of diverse racial and ethnic groups remains a common challenge in developmental research and may obscure meaningful variation in socialization processes across cultural contexts (Quintana et al., 2006). Although some research suggests general consistency in prosocial development and parenting across groups (Eisenberg et al., 2015), other work highlights culturally grounded differences (Knight et al., 2016; Wang et al., 2020). Accordingly, the present findings should be interpreted with caution, and future research with more diverse and balanced samples is needed.

Finally, the current study focused specifically on progressive parenting beliefs, best friend dyads, and prosocial behavior. Future research could extend these findings by examining whether similar cross-household spillover processes generalize to broader peer networks, other parenting beliefs and practices, and additional developmental outcomes, including academic motivation, emotional regulation, and behavioral adjustment. Such efforts would contribute to a more comprehensive understanding of how family influences operate across peer dynamics within the broader social ecology of child development.

The present study provides novel evidence that progressive parenting beliefs can extend beyond one's own child to shape prosocial behavior within peer networks. Specifically, best friend mothers' progressive parenting beliefs were associated with children's prosocial behavior indirectly through best friends' prosocial behavior. Importantly, this indirect pathway was more evident among children from lower- and average-SES backgrounds than among children from higher-SES backgrounds, suggesting that positive developmental influences from peers and their families may be particularly consequential for children growing up in disadvantaged circumstances.

These findings highlight that parenting influences are not confined to the home but may extend across children's social networks through close friendships. By identifying a cross-household pathway of socialization through best friend relationships, this study advances socio-ecological models of development and shows how family processes can influence children beyond the immediate household.

More broadly, the findings suggest that positive developmental resources may diffuse through children's social relationships and that these benefits may be especially important for children with fewer socioeconomic advantages. In this sense, supportive parenting beliefs may generate ripple effects that extend beyond one family and contribute to the social development of other children within peer networks. The more pronounced indirect pathway observed among children from lower-SES backgrounds further suggests that cross-household socialization may function as a compensatory process, providing an additional route through which positive developmental influences can reach children facing greater contextual challenges.

In sum, these findings underscore that children's development is best understood as embedded within interconnected family and peer systems, shaped not by parents alone but by the wider community of relationships that surrounds them. Understanding how progressive parenting beliefs extend across children's social networks may help inform efforts to promote prosocial development and reduce disparities in opportunity. Such efforts may matter most for children growing up with fewer resources, for whom support reaching across households may be least available elsewhere.