This chapter explores the gaps in data that hinder the effective monitoring of children's activities and engagement in digital spaces, including the accurate measurement of screen time and its relationship with child well-being outcomes. It examines how data sources could be leveraged or integrated to address these shortcomings. Additionally, the chapter highlights the importance of incorporating perspectives from health professionals, educators, parents, and children to better understand the challenges as experienced by children and those responsible for their guidance. This approach aims to inform the development of effective policies and supports.
How's Life for Children in the Digital Age?
Abstract
Enhancing data and statistics on the impact of digital transformation on people's well-being is a key focus of countries' digital policies and the roadmap outlined by the OECD to support this transformation (OECD, 2022[1]), Children should be included in this programme as well. Data and indicators on children's digital resources, practices, and experiences in the digital environment are essential for governments and other stakeholders to comprehend the challenges associated with the increasing role of digital technologies in children's lives. To address this need, the Better Internet for Kids’ “Knowledge Hub”, established in Europe, serves as a central resource for information, evidence, and policy insights on the impact of digital transformation on children and young people in the EU, Iceland, and Norway. It supports stakeholders in monitoring policies in this area through updated data, reports, maps, indices, and country profiles, facilitated by national policy experts.
Indicator sets are valuable tools to raise awareness and foster a shared understanding of child well-being issues (Dirwan and Thévenon, 2023[2]). They provide the necessary knowledge base for governments to establish coherent goals and policy priorities, thereby promoting strategic alignment and cooperation across departments and agencies. In this context, collecting data on children in the digital environment is essential for understanding the relationships between digital activities and child well-being (OECD, 2021[3]). As a result, data collection is increasing in many countries. However, assessments of ongoing initiatives suggest that data collection in this domain remains uneven and fragmented (OECD, 2020[4]). For instance, the report on Better Internet for Kids (2024[5]) highlights that only a few European countries conduct a regular (annual or bi-annual) nationally representative survey focusing on children’s digital activity, while some others collect data within the context of broader surveys. However, many countries collect data on children’s digital activities irregularly, on an ad hoc basis, or do not collect data at all.
Amidst this fragmented information landscape, internationally conducted surveys offer a valuable information base for documenting child digital well-being and the benefits and risks associated with children's engagement in the digital world. These surveys shed light on numerous issues related to child digital well-being; however, the information remains incomplete, resulting in significant important information gaps. At the international level, the review of available data in the second chapter of this report showed that the data only partially covers the dimensions necessary to fully assess children’s well-being in the digital environment. Most of the available information primarily targets teenagers, emphasises risks and negative online experiences, and includes significant details on the different types of activities children engage in when using digital devices. However, these data provide limited insights into children’s positive experiences in digital spaces, as well as on the support protections and boundaries set in their home and school environments.
As discussed in Chapter 2, recent waves of the PISA and HBSC surveys include data on digital skills and teenagers' engagement in the digital world. Although this is not the primary focus of these surveys, and the information provided is therefore limited, their strength lies in their large geographical coverage and the ability to link this data to other aspects of teenagers' lives, such as educational outcomes in PISA and physical and mental health in the HBSC. The advantage of this data for monitoring children's experiences in the digital environment is that it provides child-specific information and allows for the examination of disparities based on key socio-economic characteristics, such as gender, socio-economic status, or migrant background.
Table 6.1 presents a list of indicators derived from available international data, which enable the development of a comprehensive set of metrics on key aspects of access to and use of digital devices. The selected indicators are intended to capture essential aspects of online practices, uses, and experiences that are important for child well-being and amenable to changes in child protection and support systems. This includes information on:
Internet and digital device access, with data from the PIRLS and TIMS surveys for children around age 10, and from PISA for 15-year-olds.
Time spent using digital technologies outside of school. Only the PISA data contain information on time spent with digital devices for 15-year-olds. In addition to an indicator of average time, it may be useful to include an indicator of the proportion of 15-year-olds who report spending 2 hours or more per weekday, with the 2-hour threshold justified by recommendations often suggesting that screen time be kept to less than two hours per day (see Chapter 5).
Online social interactions. Digital tools allow teenagers to maintain virtual social interactions with friends from their physical world and to form new connections with peers they meet online. PISA data enables us to estimate the proportion of 15-year-old students who talk to their friends virtually, while HBSC data provides information on 11-, 13-, and 15-year-olds who have regular online contact with friends they met online.
Negative online experiences. The 2022 PISA survey makes it possible to estimates the proportion of 15-year-olds who have encountered inappropriate, offensive, or discriminatory content, as well as the spread of personal information. The HBSC surveys for teenagers and PIRLS surveys for younger children offer data on cyberbullying victimisation.
At-risk digital practices and attitudes. Three types of indicators can be developed from the PISA 2022 survey for 15- year-olds:
i. Indicators of intensive time spent on activities like video games or social media, where there is a risk of developing a sense of dependency and excessive time engagement. Responses to the PISA 2022 questionnaire allow the identification of the percentage of 15-year-olds spending three hours or more on each of these activities during weekdays or weekends. Studies, such as Przybylski (2014[6]), suggest that excessive game engagement (over 3 hours daily) may be associated with lower life satisfaction, reduced prosocial behaviour, and increased externalising and internalising problems. Therefore, identifying the percentage of adolescents spending 3 or more hours on these activities can help highlight those at risk.
ii. Attitude indicators reflecting a strong reliance on digital tools include the need to keep them close to respond immediately to messages, feelings of nervousness or anxiety when the tools are not nearby, or discomfort when unable to reply to a message, such as during class.
iii. Indicators of at-risk or inappropriate behaviour when adolescents interact with others via digital tools, such as sharing unverified information and admitting to recent acts of cyberbullying.
Protective behaviours & environment: This category includes indicators of actions that teenagers can take to protect themselves from the detrimental effects of digital tools, such as turning off social network and app notifications, adjusting settings to protect data and privacy, comparing sources of information before sharing, and discussing the accuracy of information with peers and family members.
This information can be used to highlight international differences and to position each country relative to the average or top-performing countries. To this end, it is recommended to add this information to the OECD Child Well-Being Data Portal & Dashboard, a tool for policymakers and the public to monitor countries’ efforts to promote child well-being. This could be achieved through the development of a dedicated module featuring the indicators on child digital behaviours and experiences listed above. The module would also include user-friendly visual tools to facilitate cross-national comparisons and provide users with quicker access to relevant information. These indicators are valuable for monitoring children's lives in the digital environment because they refer to specific, measurable behaviours or experiences that can be influenced by changes in context (Box 6.1). They also align with aspects that policy initiatives may aim to address.
Smart indicators are metrics or measurements used to assess and monitor progress toward specific goals or objectives. They are designed to be:
Specific: Indicators should be specific and clearly defined, with a clear meaning and scope. This means that the indicator should be focused on a specific aspect of child digital practices or experience, rather than being too broad or vague.
Measurable: Indicators should be quantifiable and measurable, so that progress towards the goal can be tracked over time. This means that the indicator should have a clear unit of measurement, such as percentages, numbers, or rates.
Achievable: Indicators should be achievable and realistic, meaning that they can be impacted by realistic changes in the context or the impact of interventions.
Relevant: Indicators should be relevant to the goals and objectives of the program or project. This means that the indicator should be meaningful and have a clear relationship to the program or project’s intended outcomes.
Time-bound: Indicators should be time-bound, with a clear timeframe for measurement. This means that the indicator should be measured at specific points in time to track progress towards the goal.
Source: OECD (2021[7]), Measuring What Matters for Child Well-being and Policies, OECD Publishing, Paris, https://doi.org/10.1787/e82fded1-en.
Cross-country comparison
Table 6.2 offers a comparative summary overview of OECD country performance levels on each of the indicators listed in. Green or red are assigned when a country is respectively well above or well below the average for the OECD area. Yellow indicates countries around the OECD average, and light grey missing data. This helps identify which countries excel or lag in different aspects of child digital well-being as represented by the indicator set. For instance, no OECD country achieves a green score across all dimensions, indicating that none surpasses the average in every area. However, Finland stands out by combining high engagement in various digital activities with better-than-average rates for negative online experiences. Conversely, lower-than-average rates of digital activities coexist with higher-than-average occurrences of negative experiences in Türkiye.
Countries' relative-to-average performance in each dimension
Note: This figure shows country performance levels on each indicator relative to the OECD average and the performance of other OECD countries. Green cells represent countries well above (at least half a standard deviation above) the OECD average on a given indicator, and red cells well below (at least half a standard deviation below) the OECD average. Yellow cells represent countries around (within half a standard deviation either way) the OECD average. The greater the number of yellow cells, the closer the clustering of OECD countries across that indicator. Light grey cells signify missing data.
*The indicator was reverse-coded, meaning that values well below (above) the OECD average are marked in green (red).
Source: OECD Secretariat calculations based on OECD (2022[8]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html; WHO (n.d.[9]), Health Behaviour in School-aged Children (HBSC) WHO Collaborative Cross-National Survey 2021-22, https://hbsc.org/about/ and IEA (2021[10]), Progress in International Reading Literacy Study 2021, https://pirls2021.org/results.
Evaluating individual country performance
While comparing country performance to the average in each dimension is important, it is equally useful to compare with the lowest- and highest-performing countries. Figure 6.1 illustrates how an individual country, exemplified by Denmark, measures up against the lowest- and highest-scoring country in each indicator set. Longer bars indicate that the country is closer to the highest scores in the respective dimension. The first inner circle represents the minimum outcome observed among OECD countries, while the outer circle represents the maximum outcome.
Figure 6.1 reveals that Denmark ranks among the highest in terms of children’s access to digital devices, including smartphone ownership starting at age 10. While Denmark has one of the highest levels of Internet and digital device access among OECD countries (indicators in yellow), the average time spent and the proportion of adolescents using digital devices for two or more hours per day remain significantly lower than the highest levels recorded (indicators in pink). Negative online experiences are relatively infrequent compared to countries where they are most common (indicators in brown), except for 10-year-old students, who report experiencing online bullying at least monthly – placing them closer to the higher levels observed across the OECD. Finally, adolescents in Denmark generally exhibit a comparatively high occurrence of protective behaviours, except for the proportion of teenagers who turn off their notifications before sleep, which remains far below the highest levels in the OECD (indicators in dark blue).
Monitoring outcomes for Denmark in comparison to other OECD countries
Source: OECD Secretariat calculations based on OECD (2022[8]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html; WHO (n.d.[9]), Health Behaviour in School-aged Children (HBSC) WHO Collaborative Cross-National Survey 2021-22, https://hbsc.org/about/ and IEA (2021[10]), Progress in International Reading Literacy Study 2021, https://pirls2021.org/results.
The review of the child digital data landscape presented in this document points to significant gaps in the existing cross-national information, hindering the ability to provide an objective overview of children's engagement with digital tools. This limits the capacity to fully capture both the benefits and risks associated with this engagement. Further development of the data and evidence on child digital life is necessary to understand the mechanisms behind the associations between children's engagement with the digital world and its outcomes, given that research often underscores uncertainties regarding causality direction. Globally, there is a lack of data to accurately capture the time spent on screens and digital activities, the content of those activities, and to assess the impact of screen exposure and use on children of all ages. For instance, it is very difficult to find reliable, comparable, especially longitudinal data that allows for connecting screen time, digital media use, and content with well-being and mental health outcomes. Longitudinal data would also help further explore the determinants and lasting effects of the time spent and the type of engagement with various digital technologies and media. Other important data gaps include:
A lack of national and international data on children's screen exposure and use. The available evidence indicates that children are exposed to screens from a very early age,1 both in preschool environments and within their family environment. This early exposure can alter child motor and cognitive development. However, when digital devices are used by parents and carers with moderation and interactively, they can support certain aspects of child learning, such as early language development. As children reach school age, they begin to develop digital skills and use online resources for both learning and play. Comprehensive data on their screen practices, awareness of associated risks, and the guidance provided by parents and teachers is essential for understanding and improving their digital experiences.
Tracking the time children spend on screens or engaging with digital media, alongside activities like reading, physical exercise, and other leisure activities, is essential for understanding shifts in their daily use of time. It also helps assess whether the presence of digital technologies in children’s lives is increasing and potentially displacing other activities. Beyond evaluating the direct impact of screen time on well-being, such data are crucial to determining whether digital activities come at the cost of other activities more surely linked to aspects of children's well-being.
A stronger focus on data collection regarding the quality of engagement can support evidence-based policies, ensuring that digital technologies are understood and leveraged to promote well-being. This requires moving beyond simplistic screen-time metrics, which overlook the key factors that determine whether digital engagement supports or hinders children's lives. Understanding how and why children engage with digital media – and how it integrates into their lives – can provide deeper insight into its impact on well-being. This is essential for moving beyond simplistic narratives that label digital technologies and media as universally "good" or "bad".
Insufficient data on the use of digital devices and services and the features that influence them. The risks associated with digital device use vary depending on the platforms, apps and services accessed. For example, social media apps can differ greatly in terms of how children socialise and consume content (Qustodio, 2020[11]). Additionally, the potential impact on children's sleep and cognitive development may be influenced by the length of screen exposure, whether the activities are passive or interactive, and the time of day when devices are used. Monitoring user activity throughout the day, as for instance made possible with app trackers, can provide better insights into usage patterns (Radesky et al., 2023[12]). This includes a better understanding of intensive or problematic use of digital services or devices, particularly smartphones, that may be visible, for instance, through how notifications are managed and children's reactions to them. More information on activities likely to lead to intensive or problematic use, such as video gaming, could be collected to better understand the different practices in this area and their association with vulnerability factors and children's well-being outcomes.
It is also crucial to identify the key features of social media services (e.g., status updates, profiles, private messages, likes) to better understand the mechanisms through which social media may impact well-being (Meier and Reinecke, 2020[13]). This can help prevent misattributing effects to the wrong causes, such as attributing them to "screen time" or a specific device rather than to a particular type of interaction. Equally important is the collection of information on the technological features of social media, video platforms, and video game services, as well as the underlying automated processes that may influence how children interact with them (e.g., the succession of short videos designed to retain users' attention or exposure to repeated, similar content).
A lack of data on the positive experiences linked to the use of digital devices and the Internet and on their benefits for children’s well-being. In the two surveys available to document adolescents’ digital activities in OECD countries, namely PISA and HBSC, the focus has primarily been on collecting information on the negative experiences and risks associated with the use of these devices. This inevitably creates a perception bias, making the digital world appear exclusively risky without fully recognising the benefits for the immediate or future well-being of children. The PISA 2022 survey includes some questions about online activities that likely foster the development of general skills, such as using digital devices to learn new things, including non-academic tasks, or to create personal content. However, a broader range of benefits and opportunities for children's subjective well-being could be considered, including whether the life online contribute to adolescents’ flourishing and self- esteem, as well as to the quality of their social relationships (Marciano and Viswanath, 2023[14]). For instance, the Global Kids Online Survey asks whether children engage in digital activities or view their online activities as contributing to community and civic participation, strengthening their personal, cultural, or religious identity, or helping them access information, networks, or services.
Children's exposure to and perception of major risks associated with digital activities is poorly documented. Currently, while the data available internationally provide information on the exposure of adolescents to certain risks, the range of risks covered is limited, and most importantly, they give little insights on children's and caregivers’ awareness of the responses, procedures, or assistance available to prevent risk exposure and respond in case of risk realisation. Several major risks could benefit from more in-depth data collection, including (i) awareness of physical and mental health risks, (ii) misinformation, media literacy and the risks induced by online social media, (iii) problematic digital behaviours related to social media and video games, (iv) risks associated with commercial content.
Limited knowledge of protective and vulnerability factors arising from children's individual circumstances and environments. Personal vulnerability factors (e.g., stress, loneliness) and protective factors beyond digital literacy (e.g., self-esteem) can affect both the use of digital services – such as the amount of time spent and the types of activities and interactions – and their subsequent impact on well-being outcomes (Meier and Reinecke, 2020[13]). Gathering information on these factors can help mitigate potential issues or assist children in building resilience.
Collecting information about the help and resources available to children within their family environment, at school, through their network of contacts and from health professionals is also critical. As indicated in Chapter 2, children's experience of the digital world is highly dependent on the resources, guidance and support they receive from parents, carers, educators, and teachers, or on the information they can exchange with or receive from their older peers. Documenting these aspects requires a holistic approach, as suggested in Chapter 2 and similar to the one used in the Global Kids Online survey. In optional modules, this survey asks parents and teachers about their mediation activities with their children to guide them in the digital world. It includes questions about their proactive involvement to help children navigate digital spaces safely, the technical support they can provide, as well as on the restrictions they put on the use of online resources (Zlamal et al., 2020[15]).
Children views, specific experiences and priorities could be given more space. To enhance children's protection against digital risks, it is essential to understand their concerns, their awareness of challenges, their perspectives on system weaknesses, and the types of protection they deem appropriate. This approach involves balancing protection with their legitimate desire to enjoy the benefits of the Internet and digital tools. To be effective, restrictions should not be imposed externally but be based on a shared understanding and assessment. Protective measures can leverage children's firsthand experiences and expertise in navigating the digital world – often surpassing that of their parents – and incorporate their insights on best practices. Additionally, it is crucial to consider the aspirations and experiences of children who are vulnerable due to disabilities, psychological challenges (e.g. neurodivergence (Coulstock, 2024[16])), sexual orientation, ethnicity, or specific family circumstances (such as those in care institutions) to ensure comprehensive protection. This inclusive approach helps mitigate specific risks they face and promotes equal opportunities by addressing inequalities carried over from the offline world into the online environment.
Diversifying vehicles of data collection
The need for data to monitor children’s digital practices and assess their impact on well-being and fill the aforementioned gaps necessitates diversifying data collection methods and using each for their specific advantages. Integrating modules focused on children's digital experiences into existing surveys is a cost-effective option that enables linking digital engagement with other dimensions covered by the surveys (e.g., academic performance in PISA, physical or mental health in the HBSC). However, the number of questions that can be added to the main questionnaire is naturally limited. Developing dedicated surveys on children's digital engagement allows for a more comprehensive examination of digital practices and how their family, school and peer environment can provide positive mediation and help mitigate risks. Nonetheless, the capacity to link these findings with learning and well-being outcomes is more restricted. Moreover, the impact of digital technology on children varies individually based on factors such as age, gender, personality, family life situation, socio-economic status, health. Therefore, it is important for surveys documenting children's digital lives to include information on these aspects.
The need for better documentation of how digital practices affect children's development and well-being, including in their early years, highlights the importance of incorporating these questions into longitudinal surveys that track children and adolescents’ outcomes, such as birth cohort studies and other longitudinal surveys involving adolescents. These surveys enable a better understanding of the temporal sequence between digital practices and well-being outcomes, identifying potential dynamics between them and the factors that mediate or mitigate the effects of digital activities on developmental and well-being outcomes.
Data from digital service providers and users’ devices are also valuable data sources, offering detailed information on how children and teenagers use these tools throughout the day. Unlike self-reported data from survey questionnaires, which can be affected by various response biases, these sources reflect actual behaviours and practices, enhancing the reliability of insights into the relationship between digital behaviours and well-being.2 Such data can provide granular insights into the time spent on different digital activities, the platforms and services used, the nature of engagement and interactions, and the behavioural responses to technical features and design elements. However, the collection and processing of this data must adhere to established data privacy and sharing regulations and ethical standards.
Observational data obtained from field-based experiments can also provide valuable insights into how children engage with digital technologies and respond to interactive designs. For instance, multi-week digital play interventions can be used to explore how certain digital services features or technologies (e.g. Virtual Reality, Video game features) influence children’s perceptions of their experiences, their reactions, their ability to interact, create, or make autonomous decisions, and ultimately, their well-being (UNICEF, 2024[17]).
Information from health and education professionals can provide valuable insights into the impact of digital practices on the well-being of children who are particularly vulnerable. Health professionals can share observations on the relationship between physical and mental health symptoms and the potential problematic use of digital services, especially in severe cases. Educators, on the other hand, offer perspectives on how children’s digital practices may affect attention spans, learning processes, and social interactions. Through their direct engagement with children and families, these professionals can identify patterns and trends, such as the influence of technology on daily routines, sleep quality, and classroom engagement, while also highlighting possible dysfunctions in children’s environments. Moreover, they play a crucial role in developing coping mechanisms and resilience strategies. Their expertise helps shape policies and programs that harness the benefits of digital technologies while addressing potential drawbacks, ensuring these tools are integrated to support, rather than hinder, child well-being.
To conclude, fostering good digital practices in children provides significant benefits for their well-being both during childhood and later in life. This report emphasizes the importance of empowering children and adolescents to harness the opportunities offered by digital technologies while effectively managing associated risks. A growing body of evidence indicates that increased digital technology use among adolescents can have positive impacts on various aspects of their development, including brain, cognitive, and social-emotional growth (Haddock et al., 2022[18]). Additionally, digital media platforms play a vital role in fostering connectedness, enabling adolescents to maintain relationships with family and friends, build new social connections, explore their interests, and access support networks (Holly et al., 2023[19]). Developing digital skills is also crucial for future professional success, as these competencies are increasingly indispensable in the workplace and will continue to gain importance in the ongoing digital transformation (OECD, 2022[20]).
For children to fully benefit from the positive aspects of this digital transformation both now and in their adult lives, they must navigate digital spaces that ensure their safety and be supported in developing digital literacy while maintaining the growth of other essential skills. They also need guidance in cultivating controlled usage habits and effectively managing the risks associated with the digital environment. Yet, effective management of digital risks also depends substantially on promoting children's offline well-being and addressing the issues that increase their vulnerability to problematic digital tool use or risky behaviour. With the increasing role of digital media in young people’s lives, it is crucial to adopt a holistic and rights-based approach to promoting their well-being both online and offline.
Monitoring trends and progress in this area in the years to come will be critical for establishing effective policies and governance institutions that support children’s safe and beneficial use of digital resources. Countries have several approaches to monitor trends: they can conduct surveys to explore children's and caregivers' usage of and attitudes towards digital tools; integrate questions on digital practices into existing surveys in other areas, such as physical and mental health education, to better understand the connections; and design surveys within an international framework to identify common challenges and address country-specific concerns; develop partnerships with digital service providers to analyse their data and get deeper insights into children’s digital practices and their relations with service design features. This chapter has demonstrated that leveraging internationally available data can help analyse the usage patterns and differences among groups of children based on their socio-economic characteristics, as well as identify risk factors or areas of concern. However, significant evidence gaps exist, indicating that countries should enhance their monitoring capabilities in this area. By choosing or combining different data collection methods, they can better inform and guide policy decisions.
Last but not least, policymakers, clinicians, teachers, parents, and young people themselves require a clear and simplified understanding of the growing body of evidence as it emerges. The process of collating, filtering, and evaluating new research findings should be guided by well-defined criteria for assessing quality, causal relationships, generalisability, and relevance to policy, education, healthcare, and social care. An explicit hierarchy of evidence could be used to inform policy decisions and practitioners, based on an assessment of the robustness, reliability, and accessibility of research evidence for practical use in decision-making or policy implementation. It implies that the evidence has been thoroughly vetted, assessed for quality, and is ready to be applied in real-world contexts, such as policymaking, education, healthcare, or social care. Assessing “evidence readiness” in the digital realm therefore can help prioritise which research findings are mature enough to inform actions or decisions, ensuring that only well-supported and relevant evidence is used in shaping policies or practices (Mansfield et al., 2025[21]).
References
[5] Better Internet for Kids (2024), “The Better Internet for Kids (BIK) Policy Monitor Report 2024”, https://www.betterinternetforkids.eu/bikmap (accessed on 13 June 2024).
[22] Boyle, S. et al. (2022), “Systematic Bias in Self-Reported Social Media Use in the Age of Platform Swinging: Implications for Studying Social Media Use in Relation to Adolescent Health Behavior”, International Journal of Environmental Research and Public Health 2022, Vol. 19, Page 9847, Vol. 19/16, p. 9847, https://doi.org/10.3390/IJERPH19169847.
[16] Coulstock, H. (2024), Bytes of advice: parenting neurodiverse children in the digital world, https://blogs.lse.ac.uk/parenting4digitalfuture/2024/12/11/bytes-of-advice-parenting-neurodiverse-children-in-the-digital-world/ (accessed on 19 December 2024).
[2] Dirwan, G. and O. Thévenon (2023), “Integrated policy making for child well-being: Common approaches and challenges ahead”, OECD Papers on Well-being and Inequalities, No. 16, OECD Publishing, Paris, https://doi.org/10.1787/1a5202af-en.
[18] Haddock, A. et al. (2022), “Positive Effects of Digital Technology Use by Adolescents: A Scoping Review of the Literature”, International Journal of Environmental Research and Public Health, Vol. 19/21, p. 14009, https://doi.org/10.3390/IJERPH192114009.
[19] Holly, L. et al. (2023), “Optimising adolescent wellbeing in a digital age”, BMJ, Vol. 380, p. e068279, https://doi.org/10.1136/BMJ-2021-068279.
[10] IEA (2021), Progress in International Reading Literacy Study 2021 (PIRLS 2021), International Association for the Evaluation of Educational Achievements, https://pirls2021.org/results.
[23] Mann, S. et al. (2025), The Common Sense Census: Media Use by Kids Zero to Eight, Common Sense Media, San Francisco, https://www.commonsensemedia.org/research/the-2025-common-sense-census-media-use-by-kids-zero-to-eight (accessed on 6 March 2025).
[21] Mansfield, K. et al. (2025), “From social media to artificial intelligence: improving research on digital harms in youth.”, The Lancet. Child & adolescent health, Vol. 0/0, https://doi.org/10.1016/S2352-4642(24)00332-8/ATTACHMENT/353A9179-5D44-47B3-94E0-A9E2986E5C64/MMC1.PDF.
[14] Marciano, L. and K. Viswanath (2023), “Social media use and adolescents’ well-being: A note on flourishing”, Frontiers in Psychology, Vol. 14, p. 1092109, https://doi.org/10.3389/FPSYG.2023.1092109/BIBTEX.
[13] Meier, A. and L. Reinecke (2020), “Computer-Mediated Communication, Social Media, and Mental Health: A Conceptual and Empirical Meta-Review”, Communication Research, Vol. 48/8, pp. 1182-1209, https://doi.org/10.1177/0093650220958224.
[8] OECD (2022), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
[20] OECD (2022), Skills for the Digital Transition: Assessing Recent Trends Using Big Data, OECD Publishing, Paris, https://doi.org/10.1787/38c36777-en.
[1] OECD (2022), “The OECD Going Digital Measurement Roadmap”, OECD Digital Economy Papers, No. 328, OECD Publishing, Paris, https://doi.org/10.1787/bd10100f-en.
[7] OECD (2021), Measuring What Matters for Child Well-being and Policies, OECD Publishing, Paris, https://doi.org/10.1787/e82fded1-en.
[3] OECD (2021), OECD Recommendation on Children in the Digital Environment, https://legalinstruments.oecd.org/en/instruments/OECD-LEGAL-0389 (accessed on 12 June 2024).
[4] OECD (2020), Protecting Children Online: An Overview of Recent Developments in Legal Frameworks and Policies, OECD Publishing, Paris, https://doi.org/10.1787/9e0e49a9-en.
[24] Parry, D. et al. (2021), “A systematic review and meta-analysis of discrepancies between logged and self-reported digital media use”, Nature Human Behaviour, Vol. 5/11, pp. 1535-1547, https://doi.org/10.1038/s41562-021-01117-5.
[6] Przybylski, A. (2014), “Electronic Gaming and Psychosocial Adjustment”, Pediatrics, Vol. 134/3, pp. e716-e722, https://doi.org/10.1542/PEDS.2013-4021.
[11] Qustodio (2020), Connected More than Ever, https://qweb.cdn.prismic.io/qweb/e59c2e0f-ef4f-4598-b330-10c430e2ec71_Qustodio+2020+Annual+Report+on+Children%27s+Digital+Habits.pdf (accessed on 30 January 2024).
[12] Radesky, J. et al. (2023), Constant Companion: A Week in the Life of a Young Person’s Smartphone Use, Common Sense Media, San Fransisco, https://www.commonsensemedia.org/sites/default/files/research/report/2023-cs-smartphone-research-report_final-for-web.pdf (accessed on 23 May 2024).
[17] UNICEF (2024), Digital Technology, Play, and Child Well-Being. Responsible Innovation in Technology for Children, UNICEF Innocenti Centre, https://www.unicef.org/innocenti/reports/responsible-innovation-technology-children#download (accessed on 18 June 2024).
[9] WHO (n.d.), Health Behaviour in School-aged Children (HBSC) cross-national survey, World Health Organization, Regional Office for Europe, https://hbsc.org/about/.
[15] Zlamal, R. et al. (2020), “EU Kids Online 2020: technical report”, https://doi.org/10.21953/LSE.04DR94MATPY7.
← 1. For instance, a 2024 survey conducted across the United States found that caregivers reported children aged 8 and younger spend an average of 2 hours and 27 minutes on screens daily (Mann et al., 2025[23]). Screen time varies significantly by age: children under 2 years old average 1 hour and 3 minutes per day, while 2- to 4-year-olds spend approximately 2 hours and 8 minutes. Those aged 5 to 8 use screens for about 3.5 hours daily. Compared to 2020, children now watch less live television and cable but spend more time viewing short videos on platforms like TikTok, Instagram Reels, and YouTube Shorts, with average daily use increasing from 1 minute in early 2020 to 14 minutes in 2024. Time spent gaming has risen from 23 minutes to 38 minutes per day, while video chatting has increased from 1% daily usage in 2017 and 2020 to 4% in 2024.
← 2. Survey questionnaires on digital services and devices are often subject to biases that can compromise data accuracy and reliability. Key examples include response bias, where respondents provide socially acceptable rather than truthful answers; coverage bias, which occurs when certain groups, such as those without access to certain digital services, platforms or devices, are excluded, leading potentially to skewed results. Other personal and environmental factors can introduce bias in self-reported data on the use of digital services. For instance, there is evidence that adolescents tend to significantly overestimate the average daily time they spend on individual social media platforms during a given period (Boyle et al., 2022[22]). Moreover, the accuracy of their self-reported usage is systematically influenced by the specific platforms involved, the participant's sex, and the total number of platforms they regularly use. These factors introduce biases into survey-based studies exploring the health-related impacts of social media use. In a meta-analysis of 106 studies, Parry et al. (2021[24]) found that self-reported media use only moderately aligns with logged data, with self-reports often failing to accurately reflect actual media usage. Additionally, measures of problematic media use exhibit an even weaker correlation with usage logs.