Content and activity that is harmful to children within scope of the Online Safety Bill

The National Centre for Social Research (NatCen) and City, University of London were commissioned by the Department for Digital, Culture, Media and Sport (DCMS) to carry out a rapid evidence assessment (REA) to provide a synthesis of evidence on content and activity that is harmful to children on services within scope of the Online Safety Bill.

About the study

The purpose of this REA was to provide an overview of available evidence on the risks to children from harmful content and activity on services in scope of the Online Safety Bill. This included a synthesis of evidence on the definition, prevalence and impacts of harmful content and activity, as well as any variation amongst different groups of children. The review focused on harmful content and activity, the scope of which was guided by the harms listed in the Online Harms White Paper. This included cyberbullying, pornography, violent content, pro-self-harm content, pro-suicide content, and content which could give rise to eating disorders. It also focused on emerging or lesser researched harmful content and activity.

Findings

• Overall, this review found that while there is a high-volume of research focused on online harms generally (as evidenced by high returns at the initial literature searching stage), much of that research falls out of the specific scope of this review. This is for several reasons. Firstly, much of the research is international and does not provide or disaggregate UK data. Secondly, there is a lack of primary research that focuses specifically on the prevalence and impacts of content and activity on children and young people in the UK. This is largely due to ethical challenges of conducting research with children and young people across much of the content and activity in scope of this review.

• This review also identified an uneven evidence base, with a considerable amount of evidence on cyberbullying and pornography, and much less on online content promoting eating disorders, self-harm and suicide and other online harms. The evidence base also has several methodological limitations. These include a lack of consistency in the definition of content and activity being investigated; a lack of consistency in the measurement of prevalence and impact; variable definitions of children and young people; a lack of distinction regarding the platforms under investigation; and frequent coverage of illegal content and activity, which is out of the scope of this review

• The findings of this review suggest several evidence gaps and opportunities for future research. Particular types of content and activity need much more thorough investigation. This includes self-harm and suicide-related content and content that could give rise to eating disorders, as well as the full-range of emerging harms, which provide very limited UK-specific data.

• As it currently stands, most research on the impacts of online content and activity is cross-sectional and relatively small-scale. Longitudinal, representative research that uses consistent impact measurements over-time is therefore required in order to help establish causal links between content and activity and harm.

• A significant evidence gap identified by this review is research that explores the nature and impacts of harmful content and activity across different groups of children. In particular, research on the experiences of children on the basis of disability, ethnicity, religion, sexual orientation, social background and other vulnerabilities is virtually non-existent, and in need of further exploration.

Methodology

The review was undertaken using a rapid evidence assessment (REA) design. This comprised of four key stages:

1. Literature searching and screening stage to identify the nature, availability and range of evidence relevant to the review.
2. Supplementary literature searching and screening to identify the nature, availability and range of evidence on lesser researched and emerging content and activity. This was supplemented through a review of relevant platforms’ policies on content regulation and the prevention of harm.
3. Critical evaluation stage to evaluate the quality of evidence.
4. Extraction and synthesis stage to extract and summarise data thematically.

Due to the limited timeframe in which this review was conducted, strict criteria were used to prioritise evidence for inclusion. This included a focus on:

• evidence published from 2011 onwards,
• studies which included UK data,
• peer-reviewed academic studies that are published or are in print, with inclusion of grey literature in subject areas with a more limited evidence base, and evidence available in the English language only