13. The pending agenda

The COVID-19 pandemic has reminded us that schools are not just places of academic learning. They are part of the social fabric of our communities, and they are increasingly expected to help children––especially the most vulnerable––thrive academically, physically and psychologically. Balancing these different elements has long been a challenge; doing so well in the digital world even more so.

One of the challenges is that while education is increasingly playing a role in supporting and developing children on these various outcomes, much of the specific expertise lies outside of the education sector. In the case of physical health and well-being, the medical profession and associated ministries and professional bodies have long taken the lead. In the case of digital technologies, expertise is often concentrated in private companies, and ministries of science and technology on the public policy front. This is not problematic per se: in fact, it would be unrealistic (and inefficient) to expect the education sector to take on these elements in addition to their traditional responsibilities. However, it does create two separate challenges for the education sector 1) forging the connections and partnerships required to access the relevant expertise and knowledge from other sectors; and 2) understanding how best to design and deliver policy and practice informed by a research base which may not address the primary concerns of education and educators.

The difficulty in connecting research to policy and practice is not specific to education. Particularly when looking at a complex intersection of domains, such as education, medicine, media studies, neuroscience, economics, sociology and psychology to name just a few, there is a lack of connection among the various research disciplines. Importantly, there is also a paucity of work looking at multidimensional trends and outcomes within the fields themselves. This volume, along with its sister volume on emotional well-being in the digital age (Burns and Gottschalk, 2019[1]), aims to help identify some key gaps in our knowledge that should be filled as a priority to help inform educational policy and practice.

This chapter looks first at a number of cross-cutting themes that have emerged across the research reviews, OECD work with countries and discussions of this publication. It then highlights orientations for policy, research and practice. It ends with gaps in our knowledge and areas for improvement.

Throughout the chapters in this volume and the companion volume on emotional well-being (Burns and Gottschalk, 2019[1]), a number of themes emerged that cut across different areas and discussions.

Supporting children to thrive academically, physically and psychologically in a digital world requires focusing on each element as well as the intersections and interconnections between them. This requires us to address existing barriers in policy, practice and research. These include:

  • Governing authorities, from ministries to local administrations, tend to work in silos, and areas of overlapping responsibility are not always clearly delineated or aligned. Even when seated at the same table, there are difficult questions about whose voice counts, and when.

  • Schools must work in partnership with multiple other actors, and the responsibility and accountability for non-academic educational goals (e.g., well-being, social and emotional skills) is again overlapping and unclear. Forging strong and mutually beneficial partnerships with diverse actors is necessary; but the capacity to do this is not a given, particularly for smaller regions and schools.

  • Research is also defined by traditional fields and academic disciplines, influencing posts in research institutions as well as funding streams and grants. “Interdisciplinary” research is often considered less prestigious than “pure” research, relegated to specific journals and a subset of grants and awards.

In many ways our current society has been described as having zero tolerance for risk, particularly when it comes to developing minds and bodies. This has an impact on how schools function, from the design of playgrounds and physical spaces to accountability and governance structures. And indeed, no one wants children to be harmed. Yet overprotective parenting is linked to a set of negative outcomes for children, including greater anxiety and other mental health challenges (Ulferts, 2020[2]). In addition, as pointed out in Chapter 4, risky outdoor play is important not only for the development of physical skills, but also to learn how to manage and judge risk, and learn from peers as they in turn navigate their own choices. Similarly, research on digital risks demonstrates that children who are taught to manage risk (rather than risk being minimised or removed for them) are more able to identify cyber risks and know what to do and who to talk to if they are exposed.

In fact, supporting children to “manage” risk rather than parents or schools eliminating it is part of recognising children as active agents rather than victims and vulnerable individuals. However, despite the growing awareness of the importance of resilience and being able to manage risk, as well as the potential negative repercussions from over-protection, many parents and schools still prefer a zero tolerance to risk approach. Working to change this is delicate and must address two issues simultaneously: a) the actual level of risk for the child; and b) the perceived risk (both to the child as well as the parent and/or school). This last element is crucial: Perceived disapproval of others (e.g., being considered a bad parent) is a key influence of parental fear (Bennetts et al., 2018[3]). Teachers and schools also fear being blamed or criticised if an activity or policy deviates from a zero-risk approach. Designing practices and policies to help schools, teacher, parents and students learn to manage risk have to thus take both actual risk and perceived pressure/judgement into account.

These are important distinctions. Addressing the first part of the equation requires answering the question “what is the best way to build children’s skills as well as their resilience to risks, while still protecting them from harm?” This necessitates, for example, designing school policies around play spaces that include a risk assessment of the likelihood of potential injury and the trade-offs with actual harm. It would include the voices of the children and allow them to also weigh in on the level of acceptable risk. Similarly for digital security, a risk assessment of the potential exposure to inappropriate content would indicate how and when that might be most likely to happen, involve the voices of children, and flag the kinds and timing of interventions that would minimise trauma or other impact of having been exposed to harmful content. Important features for risk assessments thus include type of risk (risky outdoor play, for example, or cyber risks or risks associated with innovation and trying new things), as well as characteristics of the individual family, school or regional regulatory body.

Addressing perceived risk and disapproval/judgement of others is even more difficult, as it requires challenging the increasingly prevailing attitude that children are vulnerable and must be protected from all of life’s threats. The persistence of this attitude, despite a concurrent discussion of child and youth empowerment, speaks to how deeply it has become entrenched in our thinking. Clear frameworks, such as the dynamic risk benefit assessment outlined in Chapter 4, are necessary, but it will not be easy. No one wants to be in the position of seeming to advocate harm to children, or looking like a parent who is careless. Even the increasing popularity of “free-range parenting” is experiencing some backlash.

Assuming this attitude change is possible, making “managing risk” policies and practice work depends heavily on the skills of teachers and parents. For example, in terms of digital risks, those teachers and parents who are less confident of their own or their child’s digitals skills tend to take a more restrictive approach to risk, and vice versa (Livingstone and Third, 2017[4]). This suggests that interventions targeting the level of skills of teachers, parents and children have the potential to increase children’s resilience to risks while expanding opportunities of digital engagement. The same patterns play out in other sectors and challenges. Teacher education (both initial and continuing professional development) will need to systematically address these issues in an ongoing manner, adapting and updating along with emerging research, tools and ecosystems.

Over the last almost 30 years there has been an increase in socially prescribed expectations for perfection, exacerbated by social media and the success culture that drives it. In particular, young people report that they feel increased pressure to be perfect to gain social approval (Curran and Hill, 2017[5]). This is related but distinct from the risk-taking discussion laid out above. This volume has looked at this from three different dimensions: the physical (body image), the cognitive (use of smart drugs) and the digital (pressures from the myth of the digital native). Other dimensions could be added: there are increasingly unrealistic expectations of relationships, for example.

Many of these pressures are prevalent from a young age. Body image concerns, for example, can be measured from 3.5 years old and tend to increase throughout adolescence and early adulthood. It is thus surprising that the discussion tends to be trivialised despite potential far-reaching consequences on behaviour, academic success and relationships. There is a clear role for education here, not only in addressing these issues and challenging the pressure, but also reflecting critically on how it is contributing to the problem. From a student perspective, for example, feeling unable to take a risk and potentially make a mistake is stressful and can result in less, not more, innovation and creativity. The same is true for teachers.

So what is driving the pursuit of perfection? As Chapter 7 highlights, media, particularly social media, are an important socialisation agent. Focusing on media is useful because they are a modifiable risk factor, one that can be addressed and actioned in policy and practice. Yet social media are not the only drivers of the pursuit of perfection. As tempting as it might be to put the burden on “new” technology, the biggest driver is social comparison itself, not the tools used for that comparison. Perfectionism itself is also a driver: achieving or approaching perfection provides or reinforces social status. We live in a success culture, where ethics about what counts as perfection – and the acceptable paths to it – are changing.

This may seem like an academic argument, touching on ethics and philosophy. Yet it is also a very concrete and pressing issue for students, families and schools. Schools are a microcosm of the various societal drivers and expectations, and they are often unprepared to deal with these issues. Addressing and challenging the pressure for perfection requires a focus on multiple levels:

  • the individual: learning how to engage critically with “perfect” media messaging and images, using techniques to address stress, anxiety and mental health issues, understanding (and believing) that no one is perfect and that mistakes are normal;

  • the system: balancing creativity and innovation, which require risk-taking and the potential for mistakes, with accountability structures for students, teachers and schools;

  • the broader society: openly addressing this pursuit of perfection, already happening on social media with myth-busting and tools to show how images are enhanced, the embracing of different and diverse bodies and types of beauty, etc.

This discussion also raises uncomfortable questions about the goals of education as it too comes under increasing pressure to be perfect. Increasingly active parents are advocating for their children and insisting on certain standards and success. Who is the client: The parent, or the child? The goals and desired outcomes change depending on who you are serving (as a teacher and educator, but also as a doctor, psychologist, and all of the other actors with whom schools increasingly partner).

A similar argument can be made about empowerment. Student agency is increasingly becoming an important goal for systems, both from an instructional point of view and a governance point of view. So too is encouraging the active involvement of parents in the education of their children. Additionally, teachers and school leaders are increasingly positioned as autonomous actors, particularly in decentralised systems. Yet different forms of empowerment can be at odds with each other. Whose voice counts (the student? parent? teacher?) if they are not aligned? We will return to this discussion in the section below.

The following orientations are general in nature, as meeting challenges with policy solutions is often context-dependent. Devising a “one size fits all” response to an inherently multifaceted issue is thus neither possible nor desirable.

A number of system-wide challenges were highlighted across this volume and in the work of the companion volume (Burns and Gottschalk, 2019[1]).

There is a longlasting debate on the role of education in strengthening child well-being and health, and the relative responsibilities of families, education and other professionals and ministries. One difficulty is the intersectoral nature of these issues, which makes ownership and responsibility difficult to determine, particularly in the decentralised context of education in many countries. Although strategies for harmonised policy responses across ministries and levels of government are becoming more common, co-ordination of actors and roles requires careful attention to be effective. In addition, developing a local or even national policy response is necessary but may not be sufficient. In a borderless digital environment, responsible parties might be in another jurisdiction and enforcement options are currently limited or even non-existent.

Children are active users of digital technologies, and one of their preferred activities is information seeking, whether for socialisation, searching for health information or schoolwork. In respecting student rights and agency, education supports the development of strong critical thinking and media literacy skills that empower children to be well-informed agents of change. We have seen the impact of organised youth activism in terms of the global climate change protests. Student voices focus attention on the positive opportunities digital technologies afford (rather than risks, which is common in the current adult-centred research and policy discourse). These are just some of the ways in which empowered children and youth contribute to shaping the world they will inherit.

Children and youth tend to be early adopters of new digital technologies and they are also the most targeted group by digital software developers and platforms. Although it is often assumed that children and youth do not understand or do not care about their digital privacy, recent research indicates that they have a fluid understanding of their privacy, valuing specific elements over others and choosing when and where to reveal data about themselves. They may also sometimes choose to prioritise popularity (measured by the number of likes or shares on certain apps, for example) over privacy. In general, children and youth are becoming more critical and shrewder about what they see in the digital environment. This does not mean that they do not need protection or help with cyber risks, especially the youngest children. But it does imply that the type of help needed could be adapted and targeted better, taking children’s own agency into account. This also extends to the physical environment, where children are for example increasingly included in discussions of play space design and broader urban planning discussions, as well as the development of inclusive health and well-being education programmes.

Teachers are increasingly expected to help integrate students of different backgrounds, to be sensitive to cultural, linguistic and gender issues, to encourage tolerance and cohesion, to respond effectively to the needs of all students and at the same time to encourage them to be self-directed learners. Teachers are also expected to prepare students for the digital world and accompanying (and rapidly evolving) knowledge and skill sets. As highlighted by the COVID-19 pandemic, they are also increasingly expected to help develop healthy habits and monitor physical and emotional well-being. All of these tasks require specific knowledge, competencies and skills. As systems increasingly recognise the need to prepare teachers for this diverse set of roles, there must be greater effort to update and ensure the quality of initial teacher education and ongoing professional development. There is also a need to better connect the stages of teacher education, improve the alignment of the various stages, and ensure that high-quality and targeted support is available when most needed.

Ensuring student well-being in a digital world means that schools are increasingly expected to work in partnership with other actors. These include parents and families, but also health professionals, psychologists and law enforcement. Increasingly, they also include digital experts, cyber security professionals and programmers. Developing and maintaining partnerships with such a diverse set of actors, some of whom (for example those from the private sector) have different aims and goals, is a complex challenge. Although historically public and private partnerships have been limited in many systems, the speed of change of digital technology makes connecting to the expertise of the sector (the majority of which is concentrated in private tech firms) imperative.

This has a number of repercussions, including thinking through what this means for the protection of education as a public good and how to build capacity across the system, from the central ministry to the classroom, to continuously learn and evolve digital competencies along with technological change. In addition, as much of the directly measured digital use data and content (i.e. from user behaviour) is owned by private companies such as social media platforms and other providers, there is also a need for agreement on sharing data and measurements for policy and research purposes, as well as regulated restrictions on collecting and using such data.

Increased migration, globalisation and urbanisation are just some of the mega-trends shaping education. Delivering quality education in the 21st century requires adaptability and flexibility while still addressing sensitive topics related to national identity and values. This is a difficult and delicate conversation, and open discussion is necessary to recognise the impact of modern social, political, technological and demographic changes on schools and classrooms. It is also important to proactively address built-in systemic bias, for example through the “design for justice” initiative, which calls for ethics in design standards for digital technologies. This includes integrating diverse points of view when designing algorithms and code in an effort to reduce in-built bias. This is just one example: design justice operates in all areas, including urban development and community building more generally.

Fulfilling the promise of emancipation in pluralistic societies is a delicate issue, and striking a fair balance between all parties in a diverse society is not always possible. One example is vaccinations, and the tension between individual family choices whether to vaccinate a child or not, and the collective good of requiring vaccinations to attend school. Different systems across the OECD have responded to this in different ways. Education is on the forefront of these issues, and as ever more diverse sets of actors become involved in education policy and practice, there will be more such examples.

The involvement of multiple voices is crucial, and we must strengthen transparent and respectful communication between schools, families and communities. Reaching out to diverse groups is facilitated by the immediacy of social media. However, social media also permits the amplification of certain voices pressuring for quick action to high profile incidents. This must be balanced when designing and developing effective policies, which requires identifying what works, under which conditions and for whom. Yet monitoring and evaluation are often the weakest link in the policy cycle - many of the country policy solutions provided for the 21st Century Children Policy Questionnaire did not clearly state whether or not they had been evaluated and if they were proven to be effective. Even when stakeholders are clearly convinced of the utility of suggested changes, there may be practical barriers to implementation in terms of the time and resources required. Many of these barriers can be targeted, for example, through research literacy training for policy makers and practitioners, helping researchers to speak to non-academic audiences, and translating and disseminating research findings to relevant communities.

There is a plethora of actors working on health and child well-being in the digital age. Many use their own definition and measurement of key concepts, including such basics as “digital literacy”, “well-being”, “screen time” and “resilience”. Without agreed and shared definitions that are nuanced and holistic, we will not be able to generate the kinds of data that are required, both to build measurements of these competencies and to build capacity of teachers and parents to help develop these skills in children.

In addition, overly broad or overly narrow definitions of key terms can lead to inaccurate assumptions and the identification of trends that do not necessarily exist and are not comparable across time and contexts. As Chapter 5 points out, terms like “Internet addiction” or “gaming disorder” are generally agreed to be misleading for multiple reasons. Not only do they potentially create social stigma that is unhelpful in supporting children and young people with “problematic interactive media use”, the term obscures the growing evidence that individuals who already suffer from anxiety or depression are more prone to engage in problematic use of technology. Causality is thus difficult to distinguish, and interventions will be most effective when addressing both virtual and physical world concerns. Incorrect or misleading definitions thus not only obscure trends; they could also lead to less effective responses from policy and practice.

Surveys appear to be a common monitoring/measuring mechanism, but as self-report measures they are prone to bias. They also tend to replicate misconceptions, for example, the myth of the digital native as set out in Chapter 9. Another example is the conception of digital and technological skills as a 'hard skill', part of the suite of subjects in science, technology, engineering and mathematics. This is despite the body of research that demonstrates the importance of “soft” digital skills, used to communicate, interact and build social capital in the digital environment, and that it is these skills that make a difference in terms of generating positive outcomes from technology use.

In addition, as much of the directly measured digital use data (i.e. from user behaviour) is owned by private companies such as social media platforms and other providers, there is also a need for agreements on sharing data and measurements for research purposes. A need for a systemic approach to evidence-based policy making continues to be essential in determining policy priorities and in maximising protections that can be afforded by national policies.

Regional, national and international policy agendas can help fill gaps by selectively funding research in priority areas.

  • research on younger children (i.e. 0-8 year-olds)

  • greater emphasis on how and why children use technology, including differentiating between media type, content, delivery interface, features of the interaction

  • establishing causal links between technology use and child outcomes, and underlying mechanisms (e.g. what phenomena like “media multitasking” could mean for processes such as attention or working memory)

  • understanding the changing landscape of digital technology use and impacts on skills. For example, smartphones and their apps are now far more common than computers or tablets. What does this mean for digital skill development? For creative self-expression and empowerment?

  • understanding recovery after exposure to a cyber-risk, to identify where and how children seek help, what works in which context, and to help deliver messages on what they should they do if it happens again

  • a deeper exploration of the potential benefits associated with technology use such as social capital formation, enhanced cognition (i.e. spatial processing, working memory), physical activity, and teaching and learning processes

  • extending research frameworks to other technologies to better understand use, prevalence and impact, as well as opportunities and risks for education of emerging areas, including Internet of Things and Internet of Toys, as well as cognitive enhancement (smart drugs, brain stimulation etc.).

In terms of physical health, the available evidence is often not translated for an educational audience, and research results too often remain in their original field. There are also important knowledge gaps remaining.

  • importance of contexts (family/neighbourhoods/communities), policies and cultures in understanding health behaviours and determinants of health

  • filtering health information and where to go for help; how best to screen and identify at risk children

  • intervening with younger and younger children (many prevention programmes target pre/adolescents whereas multiple concerns emerge earlier)

  • explaining multiple outcomes and indicators (i.e. multidimensional trends and their intersections) to better understand what works, when and in which contexts. This includes both:

    • solving measurement challenges (e.g. the long term impacts of these trends); and

    • understanding how to embed intervention strategies in curricula for primary prevention purposes.

  • longitudinal studies

  • controlled experiments with representative samples from a broad set of regions and countries, better integration of different methodologies

  • working with young people

  • comparable international indicators, including trend data across time that is disaggregated by age or stage of childhood/adolescence

  • better use of existing big data (and data from apps, digital trace data), also in combination with other sources of information (administrative, self-reports etc.)

  • real world implications of outcomes in this field, as effect sizes published in studies are often small even if statistically significant. What do these results mean for the day-to-day lives of children and their peers? Does a “large” effect size translate into functional differences in a child’s daily cognition, behaviour, social relationships and educational outcomes?

  • clearly outlining the practicality of implementation in terms of costs, additional burden on teachers, and the necessary support that teachers need to carry out training and programmes to strengthen physical health and digital competence.

The dissemination of research results should be a planned and systematic process to allow for an interdisciplinary knowledge base that can better inform practice and policy. This could be fostered through the establishment of networks to stimulate dialogue and build communities among researchers themselves. This could also include creating or supporting brokerage agencies designed to provide the required links between research and practice as well as building relevant capacity both in the system and among stakeholders.

Much of the discussion in this volume has necessarily used averages to generalise across countries and systems. However, averages hide important distinctions within and between countries and systems that cannot be overlooked. Inequality in opportunities are ever-present, and often widen as individuals grow older. Disparities in families’ capacity to support their children (including getting them into good schools) continue to translate into differences in children’s achievements, both in and outside of the school. This is true for educational achievement (and performance on tests, including PISA), educational attainment (children from more affluent families are less likely to drop out of school without a diploma and are more likely to complete tertiary education), labour market integration and later life success.

The discussion on social mobility and the intergenerational advantages of education is long-standing in both research and policy worlds. Behind the science are serious (and difficult) societal questions about the relative responsibilities of schools and families. Education is not a magic solution for disadvantage, and it cannot replace the formative role of parents in child development. Strong partnerships and collaboration with families and communities can contribute to better learning environments, but they cannot do it all.

This volume took a comprehensive look at physical health and digital technologies in modern childhood, and the intersections between them. Along with its companion volume on emotional well-being (Burns and Gottschalk, 2019[1]), it identified key changes that often fall outside conventional education discourse and the challenges they could pose for education. It suggested possible solutions to these challenges, with the goal of providing research and policy options that will help countries in educating 21st century children and the opportunities and challenges they face in the modern world.

Many of these trends are a moving target, and reports such as this can become quickly outdated. The work for education systems across the OECD is to try to stay ahead of, or at least on top of, the curve. To do this, education, like all public sectors, must break down its silos and work across government departments and research disciplines. It must engage an increasingly broad variety of actors, including the private sector. It must also evolve and grow as our societies and citizens develop, anticipating change and finding preventative solutions rather than simply reacting to problems.

By analysing the available research and data from a broad range of disciplines and linking these findings to educational policy and practice, this volume explores the potential of education systems to proactively adapt and change along with our communities and children. We owe it to our children and youth to separate fact from fiction, take risks towards ambitious but achievable goals, and help support them to get the best start in life.


[3] Bennetts, S. et al. (2018), “What influences parents’ fear about children’s independent mobility? Evidence from a state-wide survey of Australian parents”, American Journal of Health Promotion, Vol. 32/3, pp. 667-676, http://dx.doi.org/10.1177/0890117117740442.

[1] Burns, T. and F. Gottschalk (eds.) (2019), Educating 21st century children: Emotional well-being in the digital age, Educational Research and Innovation, OECD Publishing, Paris, https://dx.doi.org/10.1787/b7f33425-en.

[5] Curran, T. and A. Hill (2017), “Perfectionism is increasing over time: A meta-analysis of birth cohort differences from 1989 to 2016”, Psychological Bulletin, Vol. 145/4, pp. 410-429, http://dx.doi.org/10.1037/bul0000138.

[4] Livingstone, S. and A. Third (2017), “Children and young people’s rights in the digital age: An emerging agenda”, New Media & Society, Vol. 19/5, pp. 657-670, http://dx.doi.org/10.1177/1461444816686318.

[2] Ulferts, H. (2020), “Why parenting matters for children in the 21st century: An evidence-based framework for understanding parenting and its impact on child development”, OECD Education Working Papers, No. 222, OECD Publishing, Paris, https://dx.doi.org/10.1787/129a1a59-en.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2020

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.