1. Summary
1.1 The evidence presented below addresses the following questions posed by the Inquiry: (1). What risks and benefits does increased internet usage present to children, with particular regard to: i. Social development and wellbeing, ii. Neurological, cognitive and emotional development, iii. Data security. (5). What roles can schools play in educating and supporting children in relation to the internet? (12). What more could be done by the Government? Could there be a more joined-up approach involving the collaboration of the Government with research, civil society and commerce?
1.2 The submission presents the results of two academic studies: 1. “Evidence on the extent of harms experienced by children as a result of online risks” (addressing question 1 with a focus on risks and conducted in collaboration with Dr Victoria Nash from the Oxford Internet Institute and Dr Monica Bulger). 2. “Children, Europe and the media” (addressing question 1 with a focus on benefits, question 5 and question 12). Sections 1.5 – 1.8 and 2 - 3.3 cover the first study, and sections 1.9 and 4 cover the second study.[1]
1.3 The first study investigated the research evidence base about the actual harms experienced by children as a result of Internet use. We looked at all peer-reviewed journal articles published on the topic in English between 1997 and 2012. We conducted a systematic literature review and we identified three main types of harms as outlined in 148 empirical studies: health-related harms as a result of using pro-eating disorder, self-harm or pro-suicide websites; sex-related harms such as Internet-initiated sexual abuse of minors, and cyber-bullying.[2]
1.4 The second project was a mixed-methods study conducted with 174 children, their parents and teachers from the UK and Bulgaria that explored what children know and how they feel about their own country and about Europe, and what factors influence their perceptions, and national and European identities. The study was particularly focussed on the role of the media (including the Internet) in that process as well as the interplay between social structures (class, gender, ethnicity), socialization agents (school, parents, peers), national context and individual agency.
1.5 The first study demonstrated that the evidence base about the actual harms experienced by children as a result of online use is rather thin. The majority of harm-related studies do not present evidence about actual harms. Instead they tend to discuss potential harm and risks. Harm is rarely defined as a term and is often conflated with risk. A lot of researchers (especially on a topic such as cyberbullying) rely on surveys or other self-reported measures as opposed to documentary evidence based on actual cases or more immersive qualitative studies (e.g. online ethnography or interviews).
1.6 The evidence in some areas is more conclusive than in others. Documented cases based on police, court and medical records present evidence about young people sexually abused and psychologically or physically traumatized as a result of establishing initial online contact with perpetrators, children assisted or encouraged in their suicide attempts after visiting pro-suicide forums, and adolescents feeling encouraged to self-harm or pursue their eating disorders as a result of regularly using self-harm or pro-eating disorder websites.
1.7 The number of children who have suffered extreme harm is very small in comparison with the overall number of Internet users. Vulnerable children are more at risk.
1.8 The thin evidence base is a significant issue because policy interventions, especially increased regulation, should be made on the basis of solid and documented evidence. There is a general prevalence of survey-based projects and/or textual analyses of websites as opposed to triangulated studies based on real-live case studies and/or utilizing a range of methods and resources. The Government can encourage this type of research by prioritising these types of projects via the available research funding channels (e.g. through the Research Councils).
1.9 The second study demonstrated that primary school can play a fundamental role as a political socialization agent and by teaching media literacy in general and digital literacy in particular from a very young age. Rather than focussing almost exclusively on the potential dangers of internet use, schools should do much more to educate children how to use the Internet (and other media) in a positive and educational way (and not just for homework purposes), including as a valuable resource on news, current affairs and information about other nations, cultures and ethnicities.
2. Risks and harms: the evidence base
2.1 We conducted a systematic literature review of peer-reviewed empirical studies about harms associated with Internet use by under-18s, published in English between 1997 and 2012. Our initial database search retrieved more than 4000 articles but we narrowed them down to 148 after strictly applying the criteria for inclusion: (a) empirical work, (b) published in peer-reviewed journal, (c) main focus of study was young people (aged under 18), (d) central focus on Internet use and (e) addressed incidents of harm originating from online interactions.
2.2 Three main types of harm were outlined in the literature: 1. Health-related harms. 2. Sex-related harms. 3. Cyberbullying. There is no or little evidence base for consumer-related and data security/privacy-related harms, which are frequently mentioned in policy debate.
2.3 Scale of health-related harms: 63 academic articles presented evidence about health harms related to Internet use. This category included: eating disorders such as anorexia and bulimia (30%), self-harm/self-injury (16%), suicidal thoughts and suicides (14.3%), Internet addiction/Problematic Internet Use (11.1%) and mental health issues such as depression or psychological distress (6.4%). Other harms mentioned were aggression, sexually risky behaviours, use of stimulants and alcohol, and obesity.
2.4 Severity and frequency of health-related harms: most studies presented evidence of perceived or potential harm (the majority of studies focusing on eating disorders analysed the content of “pro-ana”/”pro-mia” websites and/or were based on surveys) as opposed to “actual” harm (e.g., documented by practitioners or evidenced by case studies). Only 11% of the studies were based on case studies. 44.4% were based on surveys and 25.4% on textual analysis of websites.
2.5 The “scariest” actual harm cases reported suicide attempts or suicides, facilitated by or incited in the online space. The most extreme cases included individuals who felt pressurized to take their own lives after declaring that they would do that online, or cases when adults “assisted” young people in their suicide attempts by advising them what drugs to use or actually supplying them with these drugs.
2.6 The online forums of some of these websites (in particular the “pro-ana”/”pro-mia” websites) also provided a support network for sufferers, which in many cases was the only support network they had access to and trusted. A number of researchers agreed that some pro-eating disorder websites could be helpful but only to people who already had anorexia. Some websites posted disclaimers which said that if you were not an anorexic, it was better not to read the information on the website because it might be harmful for you. Potentially harmful content included: “thinspiration” materials, weight loss or purging techniques and prevention of help-seeking and recovery by presenting anorexia as a lifestyle as opposed to an eating disorder.
2.7 Scale of sex-related harm: 49 academic articles presented evidence about sex-related harms linked to Internet use, focusing predominantly on online solicitation/grooming, child abuse and pornography. Three main types of studies: 1. 51% analysed the process of Internet-initiated sexual abuse, frequently on the basis of offenders’ accounts. 2. 45% studied predictors of sexual offense such as risk factors and offenders’ characteristics as well as whether the consumption of child abuse images (“child pornography”) led to offline child sexual abuse. 3. 25% of the studies investigated the “effects” of exposure to pornography or being a victim of Internet-related sexual abuse such as a range of psychological and physical harms.
2.8 Severity and frequency of sex-related harms: the evidence base in this category was much stronger because a number of studies (18%) were based on police and/or medical records. Nonetheless, more than half of the articles reported survey data (some surveys were repeatedly used). The scale of harm in Internet-initiated child sexual abuse was extensive but the evidence about the link between viewing and downloading child abuse images and offline sexual abuse was inconclusive. Harms reported: victims suffering from feelings of shame, hate, disgust, fear, repression, guilt and speechlessness.
2.9 Cyberbullying: 36 academic articles presented evidence about potential or actual harms related to cyberbullying. 36% outlined the prevalence of cyberbullying, 17% investigated the impact of cyberbullying on anxiety, 11% looked at the association between depression and cyberbullying, and 11% were focused on predictors of cyberbullying. The majority (97.2%) of cyberbullying studies reported survey results.
2.10 Scale and impact of cyberbullying: studies reported significant cross-national, intra-national and age differences in the prevalence of cyberbullying – from 9% to 72%. There was a clear association between cyber-bullying/Internet harassment and psychological harms such as distress and depression as well as self-injury and suicide attempts. Victims of both cyber and traditional bullying were most likely to experience psychological issues.
2.11 Most cyberbullying studies investigating the link between cyberbullying and anonymity concluded that in the majority of cases the victim knew or suspected who the cyberbully was. Cyberbullying was linked to children’s overall social positioning and experiences and did not occur in a vacuum.
3. Internet harms: definitions
3.1 Harm was operationalized in 44.5% of the health-related harm studies. It was defined as self-harm in the majority of these studies, including parasuicide, self-mutilation or self-injury. Other types of harm mentioned were: physical harm, desensitization to violence in real life, psychological harm, encouragement of eating disorders and distraction to academic performance.
3.2 Harm was not explicitly defined in most of the sex-related harm studies. In about a third of them an assumption could be made about what the authors meant by harm, for example, sexual abuse, bodily/physical harm, emotional/psychological harm, and social harm.
3.3 Harm was explicitly defined as a term in only 27.8% of the cyberbullying studies. Half of them discussed psychological harm and negative effects, including depression. Other types of harm mentioned were: self-harm, social harm, and potential for physical harm.
4. Benefits of Internet usage and the role of school
4.1 The evidence presented in this section is based on a mixed methods study conducted with 174 9-10-year old children, their parents and teachers. The study investigated the role the mass media (including the Internet) play in relation to children’s national and European identities as well as their knowledge of Europe and the European Union. The project also investigated the interplay between the main socialization agents (media, school, parents and peers), social structures (class, gender, and ethnicity) and individual agency in the process of identity formation and political/cultural knowledge acquisition. The study included semi-structured interviews with children, parents and teachers as well as content analysis of news items, textbooks and national curricula.
4.2 Children’s political socialization as citizens is rarely studied in academic research and it is also of little (if any) interest to policy makers, but it is of extreme importance because identity formation is an active process that does not happen overnight once adolescents “officially” become adults. Political participation and civic engagement are key components of an effective democracy so it is important to trace the role of the different socialization agents (including the Internet) in that process. Furthermore, empowered and knowledgeable children are in a much better position to resist and confront extremism and bigotry, which appear to be on the rise in British society. Mediation by parents, teachers and peers is an essential aspect of children’s political socialization.
4.3 38.8% of the children in my English sample indicated that the Internet was a source of information on news for them even at that relatively young age. School was a source of information on news for 35.8% of the children and parents for 52.2%.
4.4 The acquisition of political information in primary school children is most effective when even if initially accessed via the media, it is subsequently mediated either by parents or by school (and in some cases by peers). When shown photographs of national and international symbols, historical figures and current personalities, including high-profile national and international politicians, children recognised the symbols and faces they had either seen in the media (mainly on TV) or at school. When prompted to provide more information about the respective personality/symbol, children showed much greater confidence and ability to do that in cases when they had discussed who that person/symbol was at school in addition to the information they might have come across in the media.
4.5 While TV’s role in this process of knowledge and collective identity formation seemed indisputable, it was much more difficult to establish what role the Internet played and how school could facilitate this role. Children reported engagement in a range of activities online, but acquisition of political and cultural information did not appear to be channelled through a small number of specific websites.
4.6 Schools can play a much more active role in that respect by encouraging the positive use of the Internet, including as a source of political information and by pointing children in the direction of reliable websites or in the very least teaching them how to identify reliable websites. Media literacy is essential and schools should play a much bigger role in educating children about the media than they currently do.
4.7 Children receive plentiful guidance at school on how to avoid online dangers but they need more structured and directed guidance on the positive usage they can make of the Internet, including as a source of political and cultural information. Quite indicative of the generally negative/safety-driven discourse about the Internet is the fact that most schools mark Safer Internet Day (an EU initiative) but do not celebrate International Internet Day, which has a much more positive focus. Internet safety can be much more effectively taught if we shift the focus of the narrative in a more positive direction.
4.8 School and socio-economic status are the two factors that influence children’s knowledge of and feelings towards other nations and cultures the most. Therefore, school’s role in that respect is of essential importance. Poorer children did not even realise that the UK was part of Europe (One child from a deprived area remarked: “Europe, is that in New York?”). Given the widespread availability and use of the Internet even among children in deprived areas, schools should play a much more substantial role in relation to media literacy/digital literacy and political socialization/civic engagement.
23 September 2016
5
[1] Detailed findings on these and a range of other related topics can be obtained via the University of Chester’s institutional repository at http://chesterrep.openrepository.com/cdr/simple-search?filter_field_0=author&filter_type_0=equals&filter_value_0=Slavtcheva-Petkova%2C+Vera&sort_by=dateissued&order=DESC
[2] This project was funded by the University of Oxford’s John Fell OUP Research Fund.