Introduction

The time parents spend with their children has increased steadily since the 70s (Gershuny, 2000). Analyses of time diaries from 1975 to 2000 have shown that parent time, across all social groups, has increased and that the gap between fathers’ and mothers’ time spent has narrowed (Bianchi et al 2007). In the UK, similar trends in parental involvement have emerged. Over the last decades, there has been a significant rise in the number of parents who routinely support their children’s learning in the home. In 2007, a large-scale survey on parental involvement in children’s education, commissioned by the then Department for Children, Schools and Families in England, showed a significant increase in the number of parents involved with their children’s learning during the first decade of this century, with the largest increase being in the percentage of parents who read frequently with their children reaching 79% (Peters et al 2008). In the USA, parental involvement with children’s learning (i.e., the number of parents who read daily to their children) has also been on the rise: 53% in 1993 to 60% in 2005 (Federal Interagency Forum on Child and Family Statistics 2006).

As children move into adolescence, patterns of parental involvement change. Studies have shown that parental involvement, school- based involvement in particular, tends to decline during secondary school. Learners become more autonomous in setting their own goals and handling learning tasks and are less likely to desire parental supervision of their work (Seginer 2006). As such, direct forms of parental involvement that were desired during primary school may be less relevant and effective as children move into adolescence (Park and Holloway 2013). The nature of parent-child relationships also changes. Although the relationship remains symbiotic in that parents’ behaviour is affecting and affected by child’s behaviour and dispositions, as children grow up, parent–child relationships become less hierarchical and more bidirectional, entailing new forms of interaction based on discussion and negotiation (Smetana 2011). The repositioning of the relationship between parents and adolescent learners is likely to shift patterns of parental learning support. Hill and Tyson (2009) coined the term ‘academic socialization’ to describe the forms of parental involvement that are typical during secondary school. Academic socialization refers to interactions and conversations about academic aspirations and expectations of academic achievement, learning strategies, negotiating learning structures and goals and considering plans for the future. As a form of parental involvement, academic socialization is more subtle and responsive to adolescents’ developmental needs for autonomy, and more strongly linked to achievement than was home- or school-based involvement (Hill & Tyson, 2009).

Research on the association between parents’ socio- demographic characteristics and patterns of involvement has shown that the frequency of parental learning support with young children is roughly equal across educationally and financially diverse parents (although socio-economic factors have been found to exert medium to strong effects on children’s academic achievement-see Gregg and Washbrook 2011; Hartas 2012; Sullivan et al 2010). Although maternal education and reading habits make a substantive contribution to children’s language, literacy and social development, socio-economically advantaged parents are not more likely to help their children with homework than disadvantaged parents (eg, Ho, 2009; Lee and Bowen, 2006; US Department of Education, 2006). At secondary school, mothers’ education has been found to be particularly powerful in predicting less direct forms of involvement such as parents’ educational aspirations and planning for further education (Park and Holloway, 2013). The frequency of parental involvement and parents’ willingness to support their children’s learning have not been found to differ across diverse groups of parents (Hartas, 2011, 2012). However, there is variation in the quality of involvement which may be due to the fact that educated parents are more likely to provide their children with cognitively stimulating experiences and interactions such as going on museum and theatre trips or engaging in conversations about books and contemporary social and political topics.

Although there is a large body of research on parental involvement in the UK and USA, little research exists in other OECD (Organisation for Economic Co-operation and Development) countries. Despite much research on the benefits of parental involvement, there is ambiguity with regard to the types and forms of involvement that sustain parental learning support at different school trajectories and the social ecology within which it occurs. Furthermore, most studies that utilised PISA have focused on comparative analyses of 15 year olds’ academic performance, especially in maths, science and problem solving that are particularly important for knowledge-based economies. Few studies have focused on examining patterns of parenting to compare a wide array of parental involvement to offer cross- national perspectives on parenting practices at different trajectories of students’ school life (with the exception of Borgonovi & Montt, 2012). And even fewer studies (with the exception of a study by von Otter, 2014) have made a distinction between academically and non-academically orientated parental involvement to examine whether certain forms of parental involvement are more conducive than others to enabling continuous parental support across primary and secondary school. Finally, there are relatively few large-scale studies examining parental involvement at a secondary school level. Considering that during adolescence developmental needs change dramatically, forms of parental involvement deemed to be efficient in early stages may no longer be useful or sustainable or even desirable (Park and Holloway, 2013).

The purpose of this study was two- fold: First, to delineate patterns of parental involvement with children’s education in selected OECD countries at two points in time (at the start of primary school and when students were 15 years old) and examine changes in the frequency of parental involvement as a function of parents’ education. Secondly, to examine the contribution of various factors (i.e., parental education and reading habits, parent-child interactions, parental support with emergent literacy, school-based parental support and home educational resources) to parental literacy support at age 15. Parental education was chosen as a proxy of intellectual and cultural capital in families. The intellectual capital accumulated through education influences the ways in which parents interact with their children, the type of activities they promote and the attitudes, beliefs and values they express towards learning, as well as their views about child development and the capabilities they wish to develop in their children (Hoff et al., 2002). Moreover, parents’ education is less varied over time and thus a better indicator of families’ cultural and socio-economic background.

The research questions that guided this study were:

What are the patterns of past and current parental involvement with children’s learning in 7 OECD countries?

Are there differences in the frequency of parental involvement as a function of parents’ education?

What is the cumulative and unique contribution of parents’ educational qualifications, reading habits and attitudes; parent-child interactions; active involvement with children’s literacy at primary school; and school-based involvement to parental literacy support at age 15?

Method

Sample

The data for this study came from the Programme for International Student Assessment (PISA, 2009), conducted by the OECD. PISA started in 2000 and takes place every three years. A key objective of PISA is to examine factors related to the educational attainment of 15-year olds, near the end of their compulsory education. PISA employs a two-stage stratified sampling method, i.e., the sample of students is formed by first selecting a sample of schools, and then selecting a sample of students within each of those schools. The first-stage sampling units consist of schools with 15-year-old students. PISA collects data from both OECD and non- OECD countries and offers an opportunity to study patterns of parental involvement across many countries and economies. In 2009, 14 countries and economies that participated in PISA were offered a questionnaire to be filled out by the parents of the children who took the PISA test. For the purpose of this study, the data were obtained from the questionnaire distributed to 58653 parents (mothers mainly) in 7 OECD member countries (i.e., Denmark, Germany, Hungary, Italy, Korea, New Zealand and Portugal). The PISA surveys and assessments are specifically designed and tested to ensure comparability across countries and economies. Full details about the origins and objectives of PISA can be found at http://www.oecd.org/pisa. To adjust for unequal selection probabilities, differential nonresponse, potential sampling error and for response rate differences between subgroups of the sample the data were weighted.

Measures

An exploratory factor analysis (i.e., Principal Component Analysis with varimax rotation) was conducted to identify patterns among parent questionnaire items that referred to parents’ active engagement in literacy activities with their children at the start of primary school and at age 15, parent-child conversations at age 15, parental reading habits, school choices and parental involvement at school level. The variance explained by the emerging factors was 41.6% of the total variance. The Bartlet’s test of sphericity X2(1540) = 364121, p<.001 which was highly significant and the KMO=.888 value was high, with both statistics indicating that the data were appropriate for factor analysis (Table 1).

Table 1. Factor Analysis

Note: The eigenvalues for the factor loadings were set to be greater than .4There are three sets of measures used in this study, namely parental involvement with children’s learning at the start of formal education and at age 15; school-based involvement including school choices; and parents’ educational qualifications, reading habits and education resources (eg, home-based learning resources).

Home-based parental involvement:

Measures on parental involvement with students’ literacy included parents’ involvement with literacy activities at the start of primary school and present levels of interactions / conversations with students aged 15 (Table 1). Specifically, parents were asked whether they told stories, sang songs or played with alphabet toys with their child at the start of primary school, and whether, at the time of the PISA test – that is, when their child was 15 – they engaged with their children in conversations about school and topical issues (eg, politics, films, books). Their responses were rated as ‘Never or hardly ever’; ‘Once or twice a month’; ‘Once or twice a week’; and ‘Every, or almost every day’.

School-based parental involvement:

Parents were asked about their school involvement through volunteering, assisting teachers or participating in local school government. Their responses were rated as ‘Yes’ / ‘No’. They were also asked about their considerations when choosing their children’s school. Their responses formed two clusters, i.e., academic and social / pastoral considerations, and were rated as ‘Not important’, ‘Somewhat important’, ‘Important’ and ‘Very important’ (Table 1).

Parents’ educational background:

Table 2. % of parents in OECD countries involved in educational activities at the start of primary school

Note: the frequency of the activities ranges between ‘every day’ and ‘once or twice a week’ (collapsing the 2 categories).

Table 3. % of parents in OECD countries conversing and helping with homework at age 15

Note: the frequency of the activities ranges between ‘every day’ and ‘once or twice a week’ (collapsing the 2 categories).

Measures on parents’ education included educational qualifications, reading habits and attitudes to reading for enjoyment and access to educational resources. Specifically, the measures of parents’ educational qualifications were based on the revised International Standard Classification of Education (ISCED, 1997). Levels at ISCED 3A, 3B, 3C correspond to Upper Secondary Education, ISCED 4 refers to post-secondary, not-tertiary education and ISCED 5A, 5B or 6 refers to tertiary education, including post graduate / advanced research programmes. With regard to their attitudes to reading for enjoyment (Table 1), parents were asked whether they consider reading to be a hobby or a waste of time, whether they spend time reading at home for pleasure, and whether they enjoy going to a library or a bookstore. Their responses were rated as‘Strongly Agree’, ‘Agree’, ‘Disagree’ and ‘Strongly Disagree’.

Data analytic plan

Initial descriptive statistics (i.e., crosstabs) were employed to examine associations between parents’ educational qualifications and i) parental learning support at the start of primary school and ii) parent-child interactions and conversations at age 15. Also, parental involvement with children’s learning was examined across the OECD countries where the questionnaire was distributed. Further, this study examined sources of variability in parental literacy support for 15 year olds attributable to factors related to parental learning support at the start of primary school, parental interactions and conversations with 15 year olds, school-based parental involvement, home literacy resources and parental education and reading habits. To this end, a series of stepwise multiple regression analyses were conducted (Table 6). In step 1, background variables such as parental education, reading habits and home education resources were entered to control for their contribution; in step 2, school-based involvement and choices (i.e., academic and social / pastoral considerations) and were entered. Finally, in step 3, patterns of parental involvement at home such as parents’ interactions and conversations with their 15 year olds and parental learning support at the start of primary school were entered. This allowed the examination of the cumulative and unique contributions of parenting practices, school involvement and school choices to supporting 15 year olds’ literacy when background educational factors were accounted for.

Table 4. % of parents involved in educational activities by parent education at the start of primary school

Note: the frequency of the activities ranges between ‘every day’ and ‘once or twice a week’ (collapsing the 2 categories).

Table 5. % of parents’ interactions / conversations with 15 year olds by parent education

Note: the frequency of the activities ranges between ‘every day’ and ‘once or twice a week’ (collapsing the 2 categories).

Results

Parental learning support in selected OECD countries

In examining the percentage of parents who routinely supported their children at the start of primary school and at age 15, certain patterns emerged. Across the 7 OECD countries, most parents engaged routinely (ranging from every day to once or twice a week) with their children’s education and intellectual development and this trend was evidenced at the start of formal education and at age 15. In some countries (i.e., Denmark, Germany, Hungary and New Zealand), nearly all parents who participated in PISA stated that they routinely read to their children (Table 2). High percentages of parents were also reported for other emergent literacy activities (eg, tell stories, play word games). Furthermore, a high percentage of parents reported to often converse with their 15 year olds about how they are doing at school and other topics relevant to their school life (Table 3). Equally, many parents reported that they frequently talk with their children about films, books and politics (with the exception of Korean parents). In most countries, between one third and half of parents reported to routinely help their 15 year olds with homework (14 % of Korean parents). This trend was reversed with regard to the frequency of library visits: high percentages of parents (over 90%) reported that they visit libraries less often (‘never or hardly ever’; ‘once or twice a month’) and only a small percentage reported frequent library visits.

Parents’ education and learning support

Associations between parents’ educational qualifications and the frequency with which they supported their children’s literacy were examined at the start of primary school and at age 15. Irrespective of educational qualifications, around two thirds of parents supported their children’s emergent literacy skills, ranging from every day to once /twice a week (Table 4). Parents educated at a degree level were slightly more likely to read books to their children (although the difference was statistically significant, X2(9) = 1431, p<.000, the effect size Cramer’s V=0.09 was very weak). Similar trends in parenting emerged with 15 year olds (Table 5). Across educational levels, over 90% of parents talked with their 15 year olds about school matters every day or once /twice a week. Around 50% of parents often talked about books, nearly two thirds about politics and over three quarters about films. Around a third of parents reported to help their 15 year olds with homework. Regardless of their educational qualifications, nearly all parents (around 92%) reported that they rarely visit libraries (‘never or hardly ever’ or ‘once / twice a month’). Parents with education at a primary / lower secondary level were roughly as likely as more educated parents to routinely engage with their children’s learning and intellectual development. As children moved from primary to secondary education, parents’ support became less homework focused and more supportive of adolescents’ general intellectual and social development.

Predicting parental literacy support at age 15

As Table 6 shows, the contribution of various parenting practices to literacy support at age 15 was significant when the background factors were accounted for. The base model (step 1) for predicting support with 15 year olds’ literacy from parents’ educational background accounted for a significant but relatively small portion of the variance (Adj R2=0.25). Among the background factors, parental education (β= .032, p<.001) and home education resources (β=.084, p<.001) were significant predictors with parents’ reading habits emerging as a strong predictor (β= .178, p<.001). The addition of parental considerations regarding school choices (in step 2) significantly improved the model fit (∆R2= 0.06, p<.001). Specifically, parents’ academic (β=.100, p<.001) and pastoral (β=.041, p<.001) considerations when choosing schools as well as their current school involvement (β=.053, p<.001) made a significant contribution to supporting their children’s literacy at age 15. The addition of home-based parental involvement (in step 3) improved the model fit significantly (∆R2= 0.53, p<.001). Parental literacy support at the start of primary school (β=.261, p<.001) and, particularly, parents’ conversations with their 15 year olds (β=.835, p<.000) emerged as strong predictors for parents’ continuing literacy support when background factors were accounted for.

Table 6. Standardised regression coefficients for parental literacy support at age15Parental literacy support at age 15

**p<.001

These findings paint an interesting picture with regard to the types and patterns of parental involvement with 15 year olds’ literacy. When parental education, reading habits and home education resources were accounted for, the strongest contribution to literacy support at age 15 was made by parents’ interactions and conversations with their 15 year olds about topical issues (eg, films, political issues, books) that were not directly related to homework. Although other forms of parental involvement such as learning support at the start of formal education, choosing schools and supporting schools / teachers contributed significantly, parent-child interactions and conversations emerged as the strongest predictor regarding parental literacy support at age 15.

Discussion

The purpose of this study was to examine forms and patterns of parental learning support at the start of formal education and at age 15 across educationally diverse groups of parents (mothers mainly) in selected OECD countries. It also examined the contribution of parents’ educational qualifications, reading habits and attitudes and home- and school-based parental involvement to the literacy support 15 year olds receive at home. The findings revealed that large numbers of parents in 7 OECD countries were actively and frequently involved with their children’s learning, and their involvement was roughly equal across education levels. In examining parents’ support with homework with their 15 year olds, a downward trend was observed in the frequency of support offered, especially when compared to parental support with emergent literacy at the start of primary school. This is consistent with previous research in that as children move into secondary school, direct parental homework support decreases (eg, Seginer, 2006). Rather, as the findings in this study showed, subtle forms of parent-child interactions that promote a wider culture of learning (eg, conversations with 15 year olds about topical issues) at home were more likely to sustain literacy support for 15 year olds.

Trends in parental involvement in selected OECD countries

The reported high levels of parental involvement with children’s learning, especially in countries such as Denmark, Germany, Hungary and New Zealand, are consistent with those reported in UK-based research (Hartas, 2011, 2012; Peters et al., 2008; Siraj-Blatchford, 2010). Specifically, the findings from analyses of UK national, longitudinal studies such as the Millennium Cohort Study and the Avon Longitudinal Study of Parents and Children (Gregg and Washbrook, 2011; Hartas, 2011, 2012) showed that over three quarters of parents reported to be routinely involved (daily or several times a week) with learning activities such as reading books, playing with alphabet toys, telling stories and helping their children with homework prior to and at the start of formal education and at the end of Key Stage 1 (age 7). Most importantly, and consistently with other studies (eg, Hartas, 2011, 2012; Peters, et al., 2008), the high frequency of parental involvement across the 7 OECD countries was found to be irrespective of parental education.

In examining parental practices among socially diverse parents, Lareau (2003) coined the term ‘concerted cultivation’ to describe middle class parental practices that are conducive to children’s learning. The notion of concerted cultivation was thought of as a counterpoint to the accomplishment of natural growth and catering for children’s basic needs (eg, food, shelter), typically found among working class families. In this study, roughly equal percentages of parents across educationally diverse groups routinely supported their young children’s learning through engagement in learning activities (eg, book reading, homework support, trips to library). This finding suggests that some forms of concerted cultivation may be distributed more heterogeneously than previously thought. This is further corroborated by Chin and Phillips’ study (2004) that showed that parents from different socioeconomic and educational backgrounds tend to endorse parenting practices that are not polarised along concerted cultivation and natural growth, pointing to a continuum of parenting practices. Although parental learning support is not confined amongst educated parents, the concerted cultivation offered by educated parents and parents with good reading habits is likely to translate into better achievement outcomes for their children, considering the evidence on the strong link between parental education and children’s academic achievement (eg, Gregg and Washbrook, 2011; Hartas, 2012; Sullivan et al., 2010). Educated parents are likely to offer effective cultivation (through a mix of accessing resources, educational services, parental position in social hierarchy, a general culture of learning at home and an intrinsic interest in intellectual pursuits); however, they are not more involved with their children’s learning than are less educated parents. Ultimately, the widespread practices of concerted cultivation by parents across diverse groups highlight the extent to which parents in OECD countries have internalized policies that approach parenting as a mechanism towards enhancing children’s educational opportunities and social advancement.

Moreover, the high levels of parental involvement in selected OECD countries, especially at the early stages of children’s formal education, resonate with practices of ‘intensive mothering’, a term that was originally adopted by Hays to describe a model of parenting ‘that advises mothers to expend a tremendous amount of time, energy and money in raising their children’ (1996; 8). This is a novel finding in that most research on intensive parenting has been conducted in the UK and the US with very little in other OECD countries. And much of the existing research questions the effectiveness of intensive parenting in supporting children’s learning. In a study by Cheadle, the effects of intensive parenting on children’s academic and social outcomes have been found to be modest. During the first years in formal education, children in concerted-cultivating families get modest returns to their parents’ educational investment, which appear to decrease as children age (2008).  Other studies (eg, Bernstein, 2011; Wall, 2010) and social commentators (eg, Gottlieb, 2011; Marano, 2004) have also questioned intensive parenting, raising concerns about its impact on children’s learning and wellbeing in the long run. Intensive parenting (although likely to maximise language and academic outcomes in early years especially in families with educated parents) is less conducive to supporting children to develop agency, healthy self-esteem, good social skills, resilience and emotional maturity (Hartas, 2014).

A culture of learning in the home

Interestingly, parental interactions and conversations with 15 year olds about topical issues emerged as the strongest predictor for supporting 15 year olds’ literacy. It appears that parental involvement that supports a wider culture of learning in the home, not a direct response to homework, is likely to sustain parental learning support during adolescence. This is consistent with prior analyses of PISA which showed that, across 21 OECD and non-OECD countries, increased social and cultural communication between parents and children was associated with higher levels of reading literacy (Borgonovi and Montt, 2012). Further, a meta-analysis of research on parental involvement confirms that subtle forms of involvement may be more conducive to fostering learning at home than direct homework support. The quality of parent–child communication was found to be more highly related to student achievement than more overt expressions of parental involvement with children’s education (Jaynes, 2010). 

As children enter adolescence, parental learning support is more subtle, manifested not so much through direct support with homework but through engagement with conversations and interactions, shaped by the family’s cultural and human capital. These interactions encompass a wider array of family resources, values, cultural discourses and ways of being that form the core of ‘family habitus’ (Archer et al 2014). The findings from this study drew a distinction between aspects of parental involvement (ie, helping with/ volunteering at school, parent-child conversations) that are likely to contribute to a sustained literacy support at age 15. Aspects of the ‘family habitus’ that encourage dialogic interactions and intellectual conversations between parents and children were found to be more effective in sustaining parental literacy support for 15 year olds than mere school-driven parental involvement, possibly because parent-child conversations and interactions can stimulate diverse intellectual interests in children and an understanding of the world and their place in it.

This finding is timely and practical considering that we know little about aspects of parental involvement sustained into secondary school that positively contribute to continuing parental learning support. It highlights the importance of fluidity and non- instrumentality in parent-child interactions and shifts our understanding of families from mere learning environments that are directly responsive to school demands to places where parents and children interact and converse about ideas that are meaningful and relevant to their life. Such view of parenting offers a counter narrative to dominant views of parental involvement with children’s learning as a continuation of the school curriculum and teachers’ work in class. Direct parental involvement with children’s learning and school life has contributed to the blurring of the boundaries between family and school, especially with regard to parents’ and teachers’ role in promoting children’s learning and good behaviour (eg, Bridges, 2010). What used to be distinctive about schooling and parenting was the clear boundaries between parents’ and teachers’ responsibilities regarding children’s education and well-being. The findings from this study point to the importance of encouraging and enabling parents to engage with their children’s learning in less instrumental ways by creating a culture of learning and intellectual curiosity in the home. A culture of learning in the home should not be confined to a mere transmission of literacy and numeracy skills but expand young people’s horizon and help them develop criticality, empathy and capacity for reflection (Hartas, 2014). Irrespective of their background, parents who read for enjoyment are likely to convey a sense of learning as pleasure, experimentation and intellectual exploration and encourage their children’s evolving intellectual capacities. With this in mind, it is important to stress that partnerships between home and school should not to extend school life into the home at the cost of family life and culture.

Strengths and limitations

There are strengths and limitations to this study. The PISA study was deemed appropriate to examine patterns of parenting in selected OECD countries because it offers data on a wide range of factors that are prominent when examining parental involvement such as parental education and reading habits, home learning resources and parental learning support and decision making at different trajectories in children’s education. Further strengths of the PISA analyses lie in the use of a population-based representative sample from OECD countries which enabled replication of other studies with fairly small samples to explore patterns of parental involvement and associations between parental learning support and education in selected OECD countries.

Although the conceptual and methodological contribution of PISA to the field of comparative education has been acknowledged, the validity of PISA datasets has been criticized. The criticism has been articulated in terms of the ‘ideology of a culturally indifferent world of education’ (Trohler, 2013; as cited in Takayama, 2013) that PISA is thought to promote. Dominant discourses on children’s learning and achievement across different countries do not always account for the unique political and social contexts within which schools, teachers and parents operate. In a culturally indifferent world of education, parental involvement with children’s learning is often approached in a technocratic manner by identifying ‘what works’ and enacting (or even borrowing) family policy to encourage parental involvement at a national and international level. Furthermore, the validity of the data on parental involvement and how often it takes place may be compromised due to cross-cultural comparability data. What parental learning support entails and how it is manifested are likely to differ across different cultures.

Another limitation in the PISA study was its reliance on parents’ self-completed questionnaires (mothers mainly) to obtain measures regarding the frequency of home learning, parents’ reading habits and educational qualifications due to the potential bias and also the independence of data. Also, the possibility of a discrepancy between parents’ self -reports and their actual behaviour exists and thus we need to exercise caution when we interpret the results regarding parents’ subjective views about learning support in the home and at school. The relatively small numbers of fathers who completed the questionnaire did not allow for examining fathers’ involvement in children’s education.

In considering the cross-section nature of the data and the fact that parent-child interactions are bidirectional, it is important to note that parental involvement with students’ learning affects but is also affected by students’ performance at school. Parents tend to offer more direct forms of learning / homework support to students who perform poorly (Park, 2008). Although the present study did not examine associations between parental involvement and students’ academic performance, the strong contribution of less instrumental forms of parental involvement to continuing support at age 15 may also reflect reactivity bias in that well- performing students are likely to attract less school-driven support from their parents. However, considering the large sample size, reactivity bias may not be a significant concern.

Finally, doing secondary data analyses has pros and cons. The technical expertise involved in PISA in terms of developing surveys is high, ensuring data of good quality. Also, doing secondary analyses has the benefit of being an unobtrusive process. However, this may affect the analysts’ considerations of the research context which are useful in taking a nuanced approach to data analysis and interpretation.

Further research

Although the findings from this study revealed interesting trends in parental involvement, a direct association between the frequency of parental involvement and child academic outcomes was not examined. Previous research utilising UK datasets has shown that although most parents are actively involved with their children’s learning, the effectiveness of their learning support varied along social class lines (Hartas 2012; Sullivan et al 2010).Taking away the advantages related to social class, parental learning support (even intensive parenting) is less likely to benefit children’s social and emotional development and capability building in the long term.As such, further research is needed to examine patterns of parental involvement at different educational trajectories and their associations with social class (including family income in addition to parents’ educational qualifications) and their associations with children’s educational attainment across OECD countries. Specifically, an examination of the interaction effects of parental education and parental involvement on academic achievement would be useful to tease out whether the effectiveness of parental support is higher in educated families. In an age of austerity, as class becomes ever more important as a determinant of outcomes in Western societies, more so than race or ethnicity, class differences are likely to be felt in the quality and effectiveness of parental learning support (although not in parents’ willingness and efforts to offer it). As such, it would be useful to examine parenting and social class in OECD countries, especially for secondary school learners.

Further research is also needed to understand the relatively low numbers of Korean parents involved with some aspects of their children’s learning. Specifically, around two thirds of Korean parents supported their children at the start of primary school, dropping to around 15% with regard to talking about topical issues with their children and helping them with homework. Previous analyses of PISA 2000 and 2003 showed that learners in East Asian regions share similar strengths in terms of high achievement, high aspiration and a positive disciplinary climate in school (Ho 2009). Considering these attributes, it may be that Korean parents’ involvement is expressed along high aspirations and beliefs in high-stake testing and examination and in fostering academic motivation in their children, favouring more subtle forms of involvement compared to direct homework support. The sharp rise of private tutoring in Korea since the mid 90s may also explain the low frequency of parental help with homework (Youl-Kwan Sung, 2011).

Finally, to understand parental influences on children’s learning requires a nuanced approach to addressing questions such as ‘what kind of support parents need to maximise the impact of their involvement’, ‘within which context parental involvement is effective’ and for ‘whom’. Parenting occurs within diverse socio-economic circumstances and, being symbiotic, is influenced by families’ material resources and cultural and intellectual capital but also by children’s characteristics, behaviour and attributes (eg, attitudes towards school, cognitive and linguistic abilities). The examination of the intersections of parenting, child outcomes and social class should be guided by the notion of the ‘social minimum’ to consider the influences of different forms of family capital on parenting and children’s learning. Key contributors to conversations about the ‘social minimum’ have been Amartya Sen, Martha Nussbaum and John Rawls. Whereas these contributors focus principally on the basic economic resources that people need to secure liberties and capabilities, future research should focus on the resources, including family capital, that are pre-conditions for effective parenting and child learning.

Conclusion and Implications

Parents and the family in general have long been an economic category of interest to policy makers. Considering the unprecedented focus of family policy in the UK and other OECD countries on parenting as a way of promoting children’s learning and compensating for social and economic disadvantage, the findings about large numbers of parents in 7 OECD countries who routinely support their children’s education have significant implications. First, they question the emphasis of family policy on parental involvement given that most parents are already involved with their children’s education. Perhaps, the focus of policy should be on supporting parents to enhance the quality of the learning support they already provide to their children to sustain a culture of learning across primary and secondary school years. Secondly, although more and more parents from economically diverse groups are actively involved with their children’s learning, the achievement gap between poor and wealthier children is wider than ever (Hartas 2011, 2012). This disjuncture prompts questions not so much about the limits of family policy but more about its unequivocal focus on parenting. And as the evidence from the 7 OECD countries suggests, this is not an isolated phenomenon. As neoliberal economic restructuring policies advance discourses of individual governance and self-responsibility, parents, mothers in particular, are held accountable for their children’s success or failure with regards to academic achievement (eg, Bridges 2010; Daly 2011; Robson 2010).

The findings regarding parenting trends in selected OECD countries offer a lens to view national family policy especially as is no longer geographically or politically bounded but spread globally. Although the evidence in this study points to active and frequent parental involvement with children’s education, many countries have ‘borrowed’ family policy initiatives, mainly from Anglophone countries, to encourage parental involvement as a mechanism for reducing the achievement gap between poor and wealthier children. An example of this is the implementation of a contract of honour between parents, schools and the state by the Education Minister in Chile in 2011, committing parents to a series of tasks that increase parental involvement in their children’s learning. The adoption of national parenting policies in several OECD countries highlights the emergence of a global family policy field in which family intervention focuses exclusively on parenting to support child development and life chances. This also raises questions about the role of OECD as a transnational policy actor and political authority in the construction of global family policy and the ways in which it influences social policy (Lingard and Rawolle 2011).

There is a fine line between respecting intimate family life and enabling parents and families to engage with their children’s education in ways that are culturally and socially meaningful to them. The patterns (and intensity) of parental learning support in countries such as Chile, Denmark, Germany, Hungary and New Zealand are comparable with those in the UK. Considering that parents in these countries are already involved with their children’s education, family policy should not place the onus on parents to narrow the achievement gap and improve their children’s academic outcomes (as it has largely happened in the UK) but widen the scope of family intervention. It is crucial for family policy to acknowledge that although most parents, irrespective of their socio- demographic characteristics, routinely support their children’s education, children’s academic achievement varies along social class lines. By focusing on parents as key influential factor in children’s life without accounting for families’ socio-economic background, public conversations about parenting being in decline are normalized while inequality and injustice are sidelined. To move forward, it may be worth developing education programmes that offer parents practical support and information about education choices and learning structures and opportunities for secondary school, as well as financial assistance (eg, grants) to support socio-economically disadvantaged young people to pursue further education and prepare for university studies.

For initiatives to be effective in supporting parents to support their children’s learning, they need to build capability in parents and families and, in so doing, to account for the impact of parental education and family capital on parental involvement. To understand patterns of parenting with 15 year olds across OECD countries and economies, cultural, structural / economic, and educational explanations must all be considered to avoid essentialist interpretations about the nature of parental involvement and its impact on children’s learning, especially as countries vary in the extent to which they are egalitarian. In unequal societies, parental learning support alone (no matter how frequently it is offered or whether it is sustained throughout secondary school) is not enough to narrow the gap in educational opportunities and academic attainment between poor and wealthier young people. To achieve this, political solutions are needed to promote family policy that is socially and culturally relevant and accounts for the structural constraints and affordances in parents’ life. This is particularly important as austerity measures are felt across OECD countries and economies, which may make them more susceptible to policy borrowing and quick fixes rather than widening the scope of family intervention to tackle deficit views of parenting as the panacea for all society’s ills.