Social capital pertains to resources embedded within social relationships among individuals or groups, facilitating purposeful actions (Coleman, 1988, 1993). It encompasses both objective connections and subjective relationships characterized by elements like trust, reciprocity, and positive emotions (Paxton, 1999). Social capital can manifest at multiple social levels, including households, neighborhoods, schools, and more. Research indicates that students’ academic achievements strongly relate to human, economic, and social capital across different contexts, notably within family and educational settings (Dufur et al., 2008; Parcel and Dufur, 2001; Portes, 2024; Sun, 1998). The mechanisms through which economic and human capital influence students’ academic performance both at home and school are crucial determinants of academic outcomes, with social capital playing a vital role (Bourdieu and Passeron, 1990; Coleman, 1988; Demir, 2021; Portes, 2024).

Social capital may leverage the cross-generation transfer of human capital in three scenarios. First, social capital may establish contexts for the inter-generational transmission of human and economic capital (Bassani, 2007; Bourdieu, 1986; Coleman, 1988, 1993; Dika and Singh, 2002; Rodríguez-Hernández et al., 2020). Second, parents with more human and economic resources can mobilize greater social capital (Dufur et al., 2013; Israel and Beaulieu, 2002; Parcel et al., 2010). In a third scenario, social capital, parallel to human and economic capital, significantly influences student achievement (Dufur et al., 2008; Dufur et al., 2013; Nygård and Behtoui, 2020).

Since Bourdieu’s seminal work in 1986 and Coleman’s in 1988, numerous studies have explored the impact of various facets of social capital on student achievement (Alfred and Addo, 2017; Demir, 2021; Dika and Singh, 2002; Ferguson, 2006; Mishra, 2020; Mouw, 2006; Sobba, 2019). Most researchers have aligned with Coleman’s conceptualization and measurement of social capital, with a specific focus on family social capital. Family social capital encompasses norms, social networks, and adult-child relationships beneficial for children’s development (Coleman, 1993). It represents the time, effort, resources, and attention parents invest in interacting with their children, monitoring their activities, and promoting their wellbeing (Coleman, 1988; Ferguson, 2006; Parcel and Dufur, 2001).

Smith et al. (1995) further elaborated on Coleman’s social capital concept, proposing that it includes both structural and process elements, relevant to educational outcomes (An and Western, 2019; Dika and Singh, 2002; Ferguson, 2006; Hoffmann et al., 2020; Liou and Chang, 2008; Murtaza, 2019). Structural family social capital refers to objective family characteristics, such as the presence of parents and the number of siblings, indicating opportunities for interpersonal interactions and their frequency and duration. Two-parent households, as in traditional families, possess social capital that is often lacking in other family structures, such as single-parent households (An and Western, 2019; Coleman, 1988; Lee, 2018; Lindfors et al., 2018; Murtaza, 2019; Pribesh et al., 2020; Smith et al., 1995). However, an increased number of siblings may compete for parental time and attention, potentially diluting available family resources per child, thus impacting academic achievement. Research consistently demonstrates that a higher number of siblings is generally associated with lower individual academic achievement, a relationship often explained by the resource dilution model (Byun et al., 2012; Coleman, 1988). In many East Asian countries, where Confucian values and collectivist norms emphasize strong familial bonds and intensive parental involvement, intact family structures tend to enhance the transmission of educational resources and support, thereby bolstering academic success (Tan and Fang, 2023). In contrast, although research in Western contexts also shows that students from intact families often perform better academically, the adverse effects of single-parent households are frequently mediated by socioeconomic factors and external social support (Coleman, 1988; Heyneman, 2016). Recent cross-national studies further suggest that the magnitude of the association between family structure and academic achievement varies considerably with national educational practices; for instance, in countries with comprehensive social welfare systems and supportive educational policies, the negative impact of nontraditional family structures on achievement may be mitigated (Lee and Borgonovi, 2022).

Process features are assessed through activities that represent the quality of parental involvement in their children’s lives (Israel et al., 2001). These activities encompass nurturing (e.g., assisting with homework, discussing school activities, setting high aspirations) and monitoring (e.g., supervising behavior at school and overseeing homework). While findings regarding the effects of parental monitoring activities are mixed, with some evidence suggesting a negative correlation with desirable academic outcomes (Bassani, 2006; McNeal, 1999; Tan and Fang, 2023), this may be explained by “reversed causality” or heterogenous operationalization of social capital. Many parents engage more with their teenage children in response to their academic and behavioral challenges (Gustafsson, 2007). In contrast, proactive nurturing activities and early parent-child interactions tend to transmit norms and expectations, influencing child development positively. In empirical studies, parental human capital (e.g., education) and economic capital (e.g., income) are often used as control variables due to their significant impact on student achievement (Bradley and Corwyn, 2002; Sirin, 2005; Van Ewijk and Sleegers, 2010). Moreover, parents with higher human and economic capital are generally more capable of mobilizing social capital (Bourdieu and Emanuel, 1996; Huang, 2009; Nygård and Behtoui, 2020; Teachman et al., 1997).

However, social capital is not universally beneficial. Portes (1998) identified potential negative consequences including exclusion of outsiders, excessive obligations, restrictions on individual freedom, and downward leveling norms. In educational contexts, intensive family social capital may generate stress, limit autonomy, or reproduce inequalities when families transmit disadvantageous norms (Lareau, 2003). When family norms conflict with school expectations, strong family social capital may paradoxically impede achievement. The extent to which these negative dimensions manifest likely varies by cultural context, with intensive parental involvement experienced as supportive in some societies but intrusive in others.

Most studies on social capital and educational achievement have focused on effects in national or regional settings (Cheung and Chan, 2008; Chu, 2004; Huang, 2009; Nygård and Behtoui, 2020), neglecting the cross-cultural validity of social capital theory. However, limited studies have identified cultural differences in the relationship between family social capital and educational outcomes, suggesting that social capital’s effects may not be universally applicable across diverse cultural contexts.

In many East Asian countries such as China, Korea, deeply ingrained Confucian values and collectivist cultural norms foster intensive, structured parental involvement that not only bolsters academic performance but can also heighten student stress through the transmission of high expectations (Tan and Fang, 2023; Xie and He, 2023). In contrast, Western educational practices tend to emphasize individualism and holistic development, with parental involvement generally characterized by emotional support and the promotion of autonomy, leading to academic outcomes that are more mediated by broader socioeconomic factors (OECD, 2019). Recent empirical studies have begun to unpack these complexities, demonstrating that the interplay between family social capital and cultural capital influences educational outcomes in ways that are uniquely context dependent (Tan and Fang, 2023; Xie and He, 2023). For example, research from South Korea indicates that while certain aspects of family social capital can mitigate academic stress, other elements may inadvertently intensify it, suggesting that the beneficial effects of social capital are not universally experienced (Jarvis et al., 2020).

These findings highlight the need for further systematic investigation to critically evaluate how cultural contexts shape the influence of family social capital on student achievement. However, existing research has not systematically examined which specific cultural dimensions moderate the explanatory power of family social capital across diverse national contexts. While Coleman’s (1988) framework emphasizes universal mechanisms such as norms, social control, and social closure, the empirical evidence suggests these concepts may not be equally applicable across culturally diverse settings. Bourdieu’s (1986) perspective that different forms of capital operate within field-specific logics and power structures suggests that cultural orientations may structure how family social capital translates into educational outcomes. Yet, research has not identified which specific cultural dimensions—such as individualism versus collectivism, power distance, or indulgence versus restraint—most strongly predict cross-national variations in family social capital’s effectiveness.

Understanding which cultural dimensions moderate social capital’s effectiveness requires theoretical specification. Hofstede et al.’s (2010) cultural dimensions framework identifies six key axes along which national cultures vary: Power Distance (acceptance of hierarchical authority), Individualism versus Collectivism (prioritization of individual versus group goals), Masculinity versus Femininity (emphasis on competition versus cooperation), Uncertainty Avoidance (tolerance for ambiguity), Long-Term versus Short-Term Orientation (emphasis on future versus present rewards), and Indulgence versus Restraint (gratification versus regulation of desires and impulses). Each dimension plausibly moderates how family social capital translates into educational outcomes. For instance, in collectivist societies emphasizing interdependence and family obligations, parental involvement may exert stronger influence on achievement as children internalize expectations to fulfill family aspirations (Byun et al., 2012). Similarly, in high power distance cultures where hierarchical relationships are normative, parental authority and family structure may more strongly shape educational trajectories. Conversely, in indulgent societies prioritizing personal gratification and leisure, supportive family interactions may be more closely aligned with cultural values, potentially amplifying their educational effects. These hypothesized relationships remain largely untested in comparative educational research, limiting theoretical understanding of social capital’s cultural contingency.

Furthermore, the operationalization of social capital in educational research has lacked consensus, leading to incomparable findings across studies. This lack of consensus has hindered the comparability of prior findings on the effect of social capital on school achievement. However, the diverse operationalization of social capital has led to inconsistent findings, making it challenging to draw generalizable conclusions (Bhandari and Yasunobu, 2009; Mikiewicz, 2021; Salloum et al., 2017). Therefore, there is a need for a more standardized and comprehensive operationalization of social capital to ensure that findings across studies are comparable and can contribute to a more cohesive understanding of the impact of social capital on educational outcomes.

Meta-analysis is a powerful statistical technique that synthesize the results of multiple studies on a particular topic to provide a comprehensive and more accurate assessment of the research question at hand (Bohnett et al., 2013). It goes beyond individual studies, pooling data from various sources to uncover patterns, trends, and effects that may not be evident in any single study. The value of meta-analysis lies in its ability to enhance the reliability and generalizability of research findings. Despite its advantages, it also has some disadvantages. For instance, studies included in a meta-analysis may vary in terms of design, population, or methodology, leading to heterogeneity in the data, which thus undermine the equivalence of the variable under study (Cooper et al., 2019). Besides, meta-analyses are often based on published studies, which may not represent the full spectrum of research conducted on a topic. Negative or inconclusive results are less likely to be published, potentially skewing the analysis. In some cases, meta-analysts may have limited access to data due to privacy concerns or data ownership issues (Littell et al., 2008).

To cope with these drawbacks of general meta-analyses, international data with uniform coordination under the supervision of international organizations such as the OECD can play an important role, especially when studying complex social phenomena like the effects of social capital and family socio-economic status. By incorporating data from diverse regions and populations worldwide, meta-analysts can better account for heterogeneity, improving the overall robustness of their findings. International datasets following rigorous criteria and coordinated by international organizations, minimize the bias of conceptualization, operationalization and measurements. This uniform practice in the research design and data collection highly improves the equivalence of variables. Further, international collaborations and data-sharing initiatives can provide researchers with access to extensive datasets that encompass various cultural, economic, and social contexts. This facilitates a more holistic understanding of the effects being studied.

In short, internationally coordinated data is invaluable in enhancing the quality and applicability of meta-analyses. It enables researchers to overcome the inherent limitations of this technique and delve deeper into complex research questions, such as the impact of family structure and socio-economic status on individuals and communities, with a global perspective.

International assessments, such as the Programme for International Student Assessment (PISA), provide a robust platform for exploring cross-cultural variations in social capital theory. These surveys collect high-quality, representative data from diverse educational systems spanning a wide range of socioeconomic contexts. Initiated in 2000 and conducted every 3 years, PISA evaluates 15-year-old students’ competencies essential for adult life, focusing on literacy in reading, mathematics, and science, rather than traditional academic achievement (OECD, 2012a). Each cycle prioritizes one major domain alongside two minor domains. In 2009, the focus was on reading literacy, with mathematics and science as secondary domains. Beyond literacy assessments, PISA administers questionnaires to students, schools, and families, gathering detailed insights into school processes and familial environments. This rich contextual data supports quantitative analyses of family-based social capital’s effects, complementing the influences of human and economic capital.

PISA 2009 was selected for its comprehensive Parent Questionnaire, administered in 14 economies, which provides detailed measures of family-based social capital critical for studying reading literacy. These measures, derived through factor analyses, capture aspects such as early parental support, current parental involvement, family structure, and sibling composition, aligning with Coleman’s (1988) social capital framework. While PISA 2018 also emphasizes reading literacy and included an optional Parent Questionnaire, it placed less emphasis on detailed family social capital measures, relying more on student-reported data for constructs like Economic, Social, and Cultural Status (ESCS). Additionally, PISA 2009’s post-2008 financial crisis context enriches our analysis of socioeconomic influences on family dynamics, making it uniquely suited for this study (OECD, 2013).

PISA employs item-response theory (IRT) to convert assessment outcomes into scale scores. Because individual student proficiency cannot be precisely estimated from a limited set of test items, PISA generates five plausible values per student—multiple imputed scores reflecting measurement uncertainty. This approach provides unbiased estimates of population parameters while acknowledging individual-level uncertainty (OECD, 2012a). In our HLM analyses, we estimated models separately for each plausible value and averaged results, with standard errors adjusted for imputation variability following standard procedures (OECD, 2012b). Additional background data are collected via questionnaires completed by students, school administrators, and parents.

To ensure data integrity, PISA adopts a two-stage stratified sampling approach. In the first stage, schools with 15-year-old students are systematically selected using probabilities proportional to the size (PPS) of their eligible student populations. In the second stage, up to 35 students per school are randomly sampled with equal probability; in schools with fewer than 35 eligible students, all are included, with a minimum of 20 students per school. Each economy contributes data from at least 150 schools, each with 20 or more students, where feasible. Sampling weights adjust for differential selection and participation probabilities across schools and students.

This study draws on the 2009 PISA sample from 14 economies that administered the Parent Questionnaire, emphasizing reading literacy—a domain heavily influenced by social interactions at home—as the primary outcome. Reading achievement scores, the dependent variable, were vertically equated to the PISA 2000 scale, standardized with an international mean of 500 and a standard deviation of 100. Table 1 presents the 14 participating economies along with the number of schools and students analyzed.

This study examines how family-based social capital—including family structure, number of siblings, early parental support, and current parental involvement—affects reading literacy among 15-year-old students across 14 economies, using data from PISA 2009. We analyze differences between economies by applying multilevel modeling to account for students nested within schools and meta-analysis to compare effects across nations. This approach helps identify universal and culture-specific patterns, informing educational policies to enhance student outcomes.

Figure 2 illustrates the explanatory power of family-based social capital on student reading literacy across 14 economies, measured by Snijders and Bosker’s (1999) R-squared relative to the null model (Model 00). The figure presents the total R-squared for three models: Model 01 (demographic model, including demographic factors: student-level gender, economic, social, and cultural status, immigrant status, and school-level economic, social, and cultural status), Model 02 (social capital model, including family social capital factors: family structure, number of siblings, early parental support, and current parental support), and Model 03 (comprehensive model, including both demographic and family social capital factors).

Model 01 explains a substantial proportion of variance in reading literacy, ranging from 54.9% in Hungary to 8.4% in Macao (Figure 2). Notably, economies where demographic factors have the lowest explanatory power—Macao (8.4%), Hong Kong (17.8%), and Denmark (27.4%)—are predominantly from East Asia or exhibit unique socioeconomic characteristics, suggesting regional differences in the influence of socioeconomic and demographic background.

Model 02, incorporating family social capital factors alone, shows varying gross explanatory power across economies. The R-squared for Model 02 ranges from 14.6% in New Zealand to 0.0% in Macao (Figure 2). Economies with the highest gross explanatory power for family social capital include New Zealand (14.6%), Qatar (14.5%), and Denmark (12.2%), while the lowest are Macao (0.0%), Croatia (1.9%), and Hungary (2.0%), indicating considerable cross-country variation. In most economies, the gross explanatory power of family social capital (Model 02) is substantially lower than that of demographic factors (Model 01), suggesting that demographic characteristics capture a larger portion of variance in reading literacy.

Model 03, combining demographic and family social capital factors, yields the highest R-squared values, ranging from 56.0% in Hungary to 8.8% in Macao (Figure 2). Other high-ranking economies include Germany (50.4%), Panama (40.3%), and Italy (38.3%), while the lowest are Macao (8.8%), Hong Kong (19.9%), and Korea (27.0%). The consistent ranking of economies across models suggests that demographic factors remain a dominant influence, with family social capital providing additional but variable explanatory power.

The net explanatory power of family social capital, calculated as the difference between Model 03 and Model 01 R-squared values, is modest, ranging from 0.4% in Macao (8.8%–8.4%) to 5.5% in Germany (50.4%–47.4%). Other notable net contributions include Panama (3.4%), Portugal (2.9%), and Chile (3.4%). This modest increment highlights a significant overlap between demographic and family social capital effects, indicating strong correlations between family-based social capital and student characteristics, particularly socioeconomic background, at both student and school levels. This overlap suggests that part of the demographic effect may be mediated through family processes or contexts shaped by social capital. However, family-based social capital may also independently influence student achievement alongside socioeconomic factors, warranting further investigation into its unique contribution.

We further examine the role of family-based social capital in explaining between-school inequalities in student reading literacy across 14 economies, using the intra-class correlation coefficient (ICC). The ICC quantifies educational inequality potentially arising from unequal learning experiences, selective enrolment, or both (Raudenbush and Bryk, 2002). Figure 3 compares the ICC for Models 00–03 to assess the extent to which demographic and family social capital factors explain these inequalities.

The ICC for the null model (Model 00) reveals substantial cross-country variation in between-school inequality, ranging from 67.5% in Hungary to 19.0% in Denmark (Figure 2). High inequality in Hungary (67.5%) and Germany (56.9%) reflects their tracked educational systems, where students are streamed into academic, technical, or vocational tracks often based on socioeconomic background as well as academic capacity. In contrast, lower ICC values in Denmark (19.0%), New Zealand (19.6%), and Lithuania (31.5%) suggest more equitable systems.

Model 01, incorporating demographic factors (student-level gender, economic, social, and cultural status, immigrant status, and school-level economic, social, and cultural status), reduces the ICC across all economies. The largest reductions occur in Hungary (67.5%–31.3%, a 36.2% reduction), Germany (56.9%–25.9%, a 31.0% reduction), and Chile (47.9%–22.9%, a 25.0% reduction), while the smallest reductions are in Denmark (19.0%–6.2%, a 12.8% reduction), New Zealand (19.6%–6.4%, a 13.2% reduction), and Macao (33.8%–30.6%, a 3.2% reduction) (Figure 2). This suggests that between-school inequality is closely tied to socioeconomic and demographic background, particularly in tracked systems like Hungary and Germany.

Model 02, which includes family-based social capital factors (family structure, number of siblings, early parental support, and current parental support), also reduces the ICC from Model 00, though the extent varies considerably across economies. The most significant reductions from Model 00 are in Denmark (19.0%–12.0%, a 7.0% reduction), and New Zealand (19.6%–14.6%, a 5.0% reduction), while the smallest reductions are observed in Hungary (67.5%–67.0%, a 0.5% reduction), Croatia (43.9%–43.1%, a 0.8% reduction), and Lithuania (31.5%–30.0%, a 1.5% reduction) (Figure 2). This indicates that family social capital explains a portion of between-school inequality, though its contribution varies substantially across national contexts.

Model 03, combining demographic and family social capital factors, yields the lowest ICC values, ranging from 41.7% in Qatar to 4.9% in Denmark (Figure 2). The net contribution of family social capital, calculated as the difference between Model 03 and Model 01 ICC values, is most pronounced in Germany (25.9%–22.4%, a 3.5% reduction), and Chile (22.9%–20.9%, a 2.0% reduction). However, in some economies like Qatar (40.8%–41.7%, a 0.9% increase) and Macau (30.7%–30.6%, a 0.1% reduction), the net contribution appears limited or even slightly increases inequality, highlighting the complex interplay between demographic and family social capital factors.

In summary, both the gross (Model 02 vs. Model 00) and net (Model 03 vs. Model 01) explanatory power of family-based social capital vary across economies. The substantial disparity between gross and net effects indicates significant overlap between demographic and family social capital influences, suggesting that family social capital may mediate the effects of socioeconomic background. This aligns with Bourdieu’s (1986) theory that social capital is influenced by economic and human capital, though further research is needed to disentangle these effects.

We continue to examine the effects of family-based social capital on student reading achievement, with and without controlling for socioeconomic and demographic characteristics, across 14 economies. Tables 2–5 report the results of Models 00–03, alongside a meta-analysis of effect heterogeneity using Hedges’ Q test (Hedges and Olkin, 1985).

To further investigate the sources of cross-country heterogeneity in the explanatory power of family-based social capital (H3), we examined its associations with Hofstede’s six cultural dimensions (Hofstede et al., 2010): Power Distance (PDI), Individualism versus Collectivism (IDV), Masculinity versus Femininity (MAS), Uncertainty Avoidance (UAI), Long-Term Orientation versus Short-Term Normative Orientation (LTO), and Indulgence versus Restraint (IVR). Following the procedures outlined in the research methods section, we first conducted Pearson correlation analyses between the gross explanatory power of family-based social capital (Model 02 R-squared) and the cultural dimension scores across the 14 economies.

Table 6 presents the correlation matrix. The results reveal significant correlations between Model 02 R-squared and two cultural dimensions: a strong positive correlation with IVR (r = 0.87, p < 0.01) and a moderate negative correlation with LTO (r = −0.57, p = 0.05). No significant associations were found with the other dimensions (IDV, MAS, PDI, UAI; all p > 0.05).

To identify the key predictors, we proceeded with stepwise regression analyses. First, a multiple regression model including both IVR and LTO as predictors of Model 02 R-squared yielded an R-squared of 0.768 (adjusted R-squared = 0.716).

As shown in Table 7, IVR emerged as a significant positive predictor (β = 0.002, p = 0.003), while LTO was not significant (β = 0.000, p = 0.680).

A subsequent regression model retaining only IVR as the predictor confirmed its robust significance (β = 0.002, p < 0.001), explaining 76.3% of the variance (adjusted R-squared = 0.739).

Figure 4 and Table 8 illustrate this strong positive linear association, indicating that economies scoring higher on Indulgence versus Restraint—characterized by greater emphasis on gratification of basic human desires, enjoyment of life, and leisure (Hofstede et al., 2010)—exhibit substantially greater explanatory power of family-based social capital in predicting student reading achievement. This finding suggests that cultural norms promoting indulgence amplify the role of family processes in educational outcomes, potentially through more relaxed and supportive family environments.

This cross-country analysis across 14 economies confirms that family-based social capital significantly predicts student reading achievement beyond human and economic capital (H1). The intra-class correlation coefficient (ICC) decreases from Model 00 to Model 03, with family social capital associated with between-school inequality, though the extent varies considerably. Nuclear families and early parental support positively correlate with achievement in several economies, while the number of siblings shows consistent negative effects, and current parental support exhibits predominantly negative correlations except in Korea. Hedges’ Q tests confirm significant heterogeneity in effects across economies, supporting H2 and highlighting the context-specific nature of family social capital’s influence on reading literacy. Furthermore, the gross explanatory power of family-based social capital is strongly positively associated with the cultural dimension of Indulgence versus Restraint (IVR), explaining 76.3% of cross-country variance and confirming H3. This suggests that in societies prioritizing Indulgence, family-based social capital exerts a stronger influence on academic outcomes compared to Restraint-oriented societies. Collectively, these findings challenge universalist assumptions, revealing that the role of family social capital is shaped by a complex interplay of demographic factors, educational system characteristics, and national cultural values.

Heyneman and Loxley (1983) effects indicate that the influence of family background upon student achievement are conditioned by income levels (Heyneman, 2016; Lee and Borgonovi, 2022), suggesting that the mechanism of family socio-economic background affecting student achievement vary across countries by levels of economic development. However, influence of material capital diminishes beyond a certain threshold while social capital becomes more significant (Salloum et al., 2018). Within the realm of educational effectiveness research, our exploration has illuminated a fundamental aspect often overlooked in prior research: the profound cultural dimension that underscores the impact of family social capital on educational outcomes. Our study traversed a diverse landscape of economies, each with its unique sociocultural milieu, to unearth the intricate variations in the strength, direction, and statistical significance of family social capital’s influence on reading literacy.

Across all economies examined, social capital has a substantial impact on student achievement in addition to the influence of family socio-economic status (SES). Moreover, there is evidence of an overlap between the effects of family SES and social capital, albeit with varying degrees across countries. Furthermore, there is a moderate, yet statistically insignificant correlation between the explanatory power of family SES and social capital, suggesting that the impact of financial capital and social capital are somewhat independent from each other on the country level. This implies that investing in social capital, without incurring additional financial burdens, could have a positive impact on student achievement. However, all these effects play out depending on socio-economic and cultural contexts.

Our research has also yielded nuanced insights into the multifaceted facets of family social capital. Among these insights, we observed the adverse impact of an increased number of siblings on reading literacy. While in most economies, a larger number of siblings tends to dilute family resources, creating an environment where individual attention becomes a precious commodity, there are notable exceptions. Additionally, our study shed light on the positive correlation between early parental childhood support and reading literacy. Children who receive substantial support during their formative years tend to fare better academically. Intriguingly, Korean current parental support exhibited anomalous positive effects, challenging conventional wisdom suggesting such support is merely reactive. In Korea’s collectivist context emphasizing constant parental involvement and intense academic investment, parental support functions proactively, reinforcing literacy development (Jarvis et al., 2020; Kim and Bang, 2017).

A particularly significant contribution is our cultural dimension analysis revealing that Hofstede’s Indulgence versus Restraint (IVR) dimension strongly predicts family social capital’s explanatory power. This demonstrates that in more indulgent societies—where gratification and enjoyment are culturally valued—family social capital exerts substantially greater influence than in restrained societies governed by strict social norms. This illuminates a critical insight: mechanisms through which family resources translate into educational success are fundamentally shaped by cultural contexts. In indulgent societies, parent-child interactions, family structure, and parental involvement directly influence reading development, aligning with cultural values prioritizing personal fulfillment and interpersonal warmth. Conversely, in restrained societies, educational outcomes may be more strongly governed by institutional factors and formal schooling processes.

Family social capital’s net impact varies substantially across economies and often encounters constraints. When considered alongside background factors, added explanatory power is frequently limited. This extends to reducing between-school inequality, where family social capital’s contribution ranges from substantial (Hungary, Germany) to minimal (Macao, Hong Kong). Achieving equitable outcomes requires multifaceted approaches considering socioeconomic background, school quality, societal influences, and—as our analysis demonstrates—compatibility between family-based mechanisms and cultural orientations.

Our findings challenge Coleman’s social capital theory. While Coleman emphasized norms, social control, and social closure, our results suggest these concepts may not be universally applicable in culturally diverse settings. The strong predictive power of IVR indicates Coleman’s framework requires cultural contextualization. Social capital operates within cultural contexts that define legitimate, valued, and effective family practices, aligning with Bourdieu’s (1986) perspective that capital forms operate within field-specific logics, extended by identifying IVR as structuring these dynamics across national contexts.

This research has important implications for theory, policy, and practice. First, it underscores the need for critical re-evaluation of Coleman’s social capital theory and measurement. The discovery that IVR strongly moderates family social capital’s explanatory power suggests researchers must consider cultural dimensions when theorizing about, measuring, and interpreting effects. Future frameworks should explicitly incorporate cultural orientation as a moderating factor rather than treating social capital effects as culturally invariant.

Second, findings highlight the importance of considering family social capital within broader educational contexts profoundly influenced by community and school norms. The cultural dimension analysis reinforces this by demonstrating that identical family practices may have different impacts depending on cultural context. Future research should explore complex interactions between social capital and other capital forms, effectiveness thresholds, and simultaneous influences of multiple social contexts.

Third, findings carry policy implications. In higher-indulgence societies, policies strengthening family social capital—parental engagement programs, family literacy initiatives, family structure supports—may be particularly effective levers for improving outcomes. Conversely, in restrained societies, strategies may need to focus more on institutional quality, pedagogical approaches, and school-based resources. This does not mean family engagement is unimportant in restrained societies, but policymakers should calibrate expectations and investments according to cultural context.

Furthermore, substantial heterogeneity in family social capital effects (Hedges’ Q tests, p < 0.01) cautions against direct policy transfer across cultural contexts. Interventions proven effective in high-IVR societies (New Zealand, Denmark) may not achieve similar results in lower-IVR societies (Hong Kong, Korea) without substantial cultural adaptation.

This study has several important limitations that warrant careful consideration. First, measurement limitations affect key constructs. The ESCS index has been criticized for lacking contextual sensitivity and pragmatic validity, not fully aligning with SES’s multidimensional, context-specific nature (Avvisati, 2020; Rutkowski and Rutkowski, 2013). This may undermine modeling family effects and cross-country comparisons, highlighting needs for improved SES measurement in PISA capturing diverse national and cultural contexts. Similarly, our family social capital measures, while theoretically grounded in Coleman’s (1988) framework, may not fully capture culturally specific practices or the quality of interactions beyond their frequency.

Second, data and sample limitations constrain generalizability. Reliance on decade-old data may not fully reflect current practices and evolving family dynamics in an increasingly digital, globalized world. However, PISA 2009’s unique focus on reading literacy and comprehensive family social capital measures provide valuable, relevant insights. The cultural dimension analysis is further limited by relatively small sample size (n = 14) and cross-sectional design. While IVR’s statistical significance (p < 0.001) and large effect size (R² = 0.763) are impressive, caution is warranted in generalizing. Replication with larger samples and recent data would strengthen confidence. Additionally, Hofstede’s dimensions represent national averages that may not capture within-country heterogeneity or how cultural values manifest in specific family practices.

Third, the cross-sectional design imposes fundamental threats to causal inference. Selection bias may operate if families high in social capital differ systematically on unobserved characteristics such as parental motivation, intelligence, or personality traits that independently affect student achievement. Our controls for ESCS partially address this concern but cannot eliminate unmeasured confounding. Omitted variable bias remains likely; while we control for SES, we lack measures of parental personality, time availability, cognitive ability, or child characteristics (temperament, prior achievement) that may jointly determine both family social capital and reading literacy. Reverse causality threatens interpretation, particularly for current parental support where parents may increase involvement in response to children’s academic difficulties rather than proactively fostering achievement. The predominantly negative coefficients for current support across most economies (except Korea) suggest compensatory involvement patterns. Measurement error in self-reported parental support may introduce bias through social desirability, with parents over-reporting involvement frequency.

Each threat likely biases estimates in specific directions: selection and omitted variables may inflate social capital effects if unmeasured factors positively correlate with both predictors and outcomes, while reverse causality may attenuate or reverse true effects for current support. Despite these limitations, our findings contribute valuable insights into cross-national variations in family social capital effects and their systematic relationship with cultural dimensions. Future research employing longitudinal designs with multiple measurement occasions, instrumental variable approaches exploiting exogenous policy changes, or family fixed-effects models comparing siblings would better establish causal mechanisms and strengthen confidence in the patterns we have identified.

Future research should explore several avenues. First, using recent datasets (PISA 2018 or later) to examine whether educational policy changes, digital technology use, and evolving family dynamics have altered family social capital’s influence (OECD, 2019), particularly whether the IVR relationship has persisted or changed. Second, broadening social capital operationalization to include digital social capital and community resources might provide comprehensive understanding of how familial support interacts with cultural contexts (Coleman, 1988; Tan and Fang, 2023). Third, employing longitudinal designs could clarify causal relationships and illuminate dynamic interplay between family social capital and academic performance over time, particularly regarding mediating factors and whether effects vary across cultural contexts (Dufur et al., 2013). Expanding cultural dimension analysis represents another promising direction. Future research could examine additional cultural frameworks beyond Hofstede’s dimensions or explore within-country cultural variation. Qualitative research exploring how families in high-IVR versus low-IVR societies conceptualize and enact parental involvement would provide valuable mechanistic insights into pathways through which cultural orientation shapes family social capital’s effectiveness.