ST0003
Written evidence submitted by Professor Peter Etchells
I am a Professor of Psychology and Science Communication at Bath Spa University, and I have been studying the effects of digital technology for a number of years. By way of background, I have also been communicating the science of what we do and do not know about screen time effects for over a decade. I was the lead author for a high-profile open letter1, published in the Guardian in 2017, and co-signed by 80 experts in the area, which emphasised the need for high quality research and evidence to inform screen time policies. I am also the author of an upcoming book, Unlocked: the real science of screen time (and how to spend it better), which will be published in March 2024.
Before I respond to specific terms of reference, it is important to outline three key issues regarding the current state of the scientific literature on screen time effects. These issues are crucial to critically considering any studies relevant to the call for evidence.
1) ‘Screen time’ is inconsistently defined, and is essentially a meaningless concept
Screen time is a generic term that makes intuitive sense, but offers little explanatory value in terms of behavioural and wellbeing effects. For a number of years, there have been growing calls in the research literature2 to acknowledge the fact that sheer time spent on screens is not as important as what it is that we do with that time (i.e., content) and the external environment within which that screen use sits (i.e. context). This issue is compounded by the fact that throughout the extant research literature, terms such as screens, screen time, digital technology, social media, social networking, and more are generally used interchangeably3. This makes it difficult to develop coherent theoretical frameworks within which to study specific screen effects. For instance, some studies find negative associations between screens and, say, mental health, some find positive associations, and some find no links, because they are all measuring different things. There is therefore a need to move beyond the use of screen time as a term that we assume has evidentiary value4, and instead focus on how we use screens to interact with each other, whether there are specific features of individual apps or media that might be important, and how these effects are mediated or moderated by other contextual factors.
2) Screen time research is often reactive, and therefore offers little in the way of predictive power
Some researchers5 have noted that much screen time research is ‘concern-centric’; that is, research occurs in reaction to public concerns about the potential detrimental impact of screens. In doing this, the majority of studies in the area adopt a framing in which digital technology use is considered implicitly and solely in terms of negative effects. In turn, this restricts researchers to looking at technology effects in terms of dose-response relationships; that is, attempting to quantify what exact ‘doses’ of screen use are problematic or harmful. However, such approaches necessarily reduce engagement with digital technology to a simple number, which does not capture the richness and complexity of the role it plays in modern life. As a result of this, there is currently little in the way of appropriate theoretical framework development within which to capture that nuance6. Theoretical models of screen use are important not just for improving our understanding of how screens impact wellbeing and behaviour currently, but for making meaningful and testable predictions about the impact of future technologies.
3) The evidence base is currently not of sufficient quality to fully inform policy recommendations
Much like earlier debates on the impact of video games on behaviour and mental wellbeing, the evidence base concerning screen time effects has a number of shortcomings that impact its utility when it comes to making policy recommendations. As noted above, there are currently no consensus definitions or taxonomies of screen time, and coupled with a scarcity of appropriate theoretical frameworks, this lends the research literature vulnerable to bias. To further compound the issue, the vast majority of research on screen time effects employs subjective self-report methodologies: asking groups of individuals to estimate their own screen use/phone use/social media use over a specific timescale. Recent research has shown that these sorts of subjective measures of screen time map poorly onto more objective measures of actual screen time taken from digital devices; for example, Dr Heather Shaw and colleagues7 have found that smartphone ‘addiction’ correlates positively with anxiety and depression only when looking at self-report measures, and not when using objective screen use measures. Moreover, the size of the reported effect is quadrupled when those subjective measures are used. There is therefore a need to improve the quality of the evidence base before effective policy can be implemented. To do this, there is a need to encourage high quality research that is pre-registered, replicated and which adheres to the principles of open science, which also leverages real-world, objective industry data.
Terms of reference
Term of reference 2: What is the current understanding of how screen time can support or impact children’s wellbeing and mental health, including the use of social media?
This is a non-trivial question to answer, as due in part to the above issues, findings are inconsistent and conflicting across the research literature – in fact, in some cases findings are inconsistent based on the same data. For example, a 2020 study8 using data from Monitoring the Future, a large, US-based longitudinal dataset, suggests that daily social media use was not a consistent risk factor for adolescent depression (although this varied depending on the particular subsample considered). This is inconsistent with previous research, also using Monitoring the Future data9, which concluded that screen use was clearly linked to poorer wellbeing. A 2021 meta-analysis10 of 37 studies in the area suggests that the current data we have does not convincingly support the claim that smartphones, or social media, are associated with poorer mental health, and further suggests that methodological rigour is a problem for the research area: where studies use appropriate control variables, any reported effects of screen time tend to be much smaller than those noted in studies which do not use appropriate controls.
As such, there is no clear or consistent answer as to whether screen time, regardless of how that is specifically defined, impacts mental health in either a positive or negative manner. Arguably, the problem here is that the question, as noted above, is often framed in terms of negative effects: that is, how does social media/screen time affect mental wellbeing, or vice versa. There is a growing understanding in the research literature that this is perhaps the wrong way to address the question. Instead, it might be better to consider why it is the case that some children and adolescents flourish online, but others encounter serious difficulties. In this frame of thinking, there is some evidence11 to suggest that offline risks and vulnerabilities mirror online risks, and that this might be particularly acute for children and adolescents. A review by Professor Candice Odgers12 and colleagues has suggested that although we often think about our online and offline worlds being unconnected, the majority of online threats to wellbeing find their basis in related offline factors, with cyberbullying being one such example – cross-sectional data13 from some 120,000 UK teenagers has shown, for instance, that 90% of teenagers who report being bullied online also report bullying offline. Data14 from US and Norwegian teens reports a similar overlap, which is also the case for perpetrators of bullying. More broadly, work from 201715 showed that specific types of offline difficulties that children and adolescents encounter – such as physical or mental health difficulties, or caring needs/responsibilities – can be predictive of the particular types of risks that are encountered online.
The question as to whether screen time supports or impacts childhood wellbeing therefore relies on an understanding of the support structures that are in place to help children and adolescents navigate their online worlds. For example, recent research16 that I have been involved in suggests that for teenagers, more time spent using a computer on both weekdays and weekends was associated with a small increased risk of anxiety, but that these effects were attenuated when the amount of time spent alone was taken into account. In part this is an issue of social inequality; there is some evidence17 to suggest that while children from lower income families have less access to computers or smartphones than their higher income counterparts, they spend around 1.5-2 hours longer on screen-based media during the day, and about 50% less time using those devices for homework. In a similar vein, surveys18 suggest that children from lower income households also tend to have less support in terms of mediation of their internet use (for example, having conversations with parents or caregivers about engaging with online content).
In other words, there are a wealth of both risks and benefits that online worlds can afford children and adolescents, and in some cases, those can have an appreciable impact on wellbeing. Pre-existing vulnerabilities and difficulties can lend themselves to greater levels of problems online, and there is a reciprocal effect where those online problems, in turn, can reinforce offline problems. The extent to which this impact is meaningful depends on a range of external factors, such as the quality, availability and types of support systems that children have access to.
The best way to think about screen time, then, is not in a dose-response manner (that is, considering whether, or to what extent, screen time is a causal driver of mental health or wellbeing issues). Instead, it is better to acknowledge that screen use is not a unitary or simple variable, but a complex and multifaceted component of a wider ecosystem of factors that impact mental resiliency.
Term of reference 4: How can schools and parents be better supported to manage children’s screen usage, for example, through age-related guidance? Could the Department for Education be doing more in this area?
There is little empirical evidence19 to support the idea of time-based limits for screen use at different ages, and often such recommendations are unhelpful as they do not support parents and caregivers in navigating conversations about how to effectively manage screen time. One of the best sets of guidelines around healthy screen time was published by the RCPCH in 201920, which encourages individualised conversations with children around why, when, where and how screen-based technologies are used in the household, and whether their needs are being met as a result. Similarly, there is emerging evidence21 to suggest that the way in which parent/child interactions are framed (for example, either through authoritarian approaches versus autonomy-supportive approaches) can impact how likely teenagers are to hide their technology use. There is therefore a need to develop clearer guidance to support schools and parents in navigating conversations around appropriate, reflective and beneficial digital technology use, as opposed to relying on strict limits. In developing this guidance, we need to have discussions about the presence and impact of age restrictions on various digital media. Recent research22, for instance, has suggested that Norwegian families take a pragmatic approach to video game ratings, treating them as suggestions rather than more strictly adhering to them. This is arguably at odds with their intended use, but we currently have an underdeveloped evidence based regarding adherence to these forms of digital technology age limits, and how such conversations are managed in both the home and school environments.
In acknowledging the paucity of the evidence base on video game effects, the Department for Culture, Media and Sport has, over the past few years, taken the approach of working with academics, games industry professionals and other stakeholders to develop the Video Games Research Framework, which provides a roadmap for conducting robust, policy-ready research. There is a need for similar endeavours beyond video games, and the Department for Education could be well-placed to facilitate the development of similar frameworks within the context of understanding screen time effects in children and adolescents.
Conclusion
I welcome the Department’s call to understand the science of what we currently do and do not know about the effects of digital technology use within an educational context. However, from a psychological perspective, the vast majority of studies published looking at the interplay between various types of screen use, behaviour and mental wellbeing are correlational in nature, and largely reactive to (understandable) societal concerns, as opposed to being grounded in robust theoretical and methodological frameworks. As such, it is not clear that the current level of scientific evidence that we have is of sufficient quality to develop clear lines of policy. In order to overcome this issue, there is a need for an independent broker to bring academic researchers and industry partners together to use large-scale, objective data in an ethical and robust manner, in order to answer meaningful questions about both the positive and negative impacts of various types of screen use. The Department for Education may be well placed to act in this capacity, following similar approaches taken by other departments of government.
References
September 2023