Challenging online situations reported by Italian and Portuguese children in 2018

Resumen Abstract Based on the results of the EU Kids Online survey (2018) already completed in Portugal and Italy, this article analyses two issues related to the growing dissemination of hate messages and fake news: firstly, how Italian and Portuguese children and young people aged 9-17 are coping with negative online content and conduct. Secondly, how they report their informational skills and their relationship with the news content. In comparison to previous surveys conducted in both countries (EU Kids Online 2010, Net Children Go Mobile, 2014), the results reveal an increase in negative online experiences – such as exposure to hate messages, to violent and gory images, or being the victim or perpetrator of cyberbullying. Although the levels of troubling experiences are much higher in Portugal, Italian and Portuguese children present relatively similar patterns of coping. In both countries informational skills relating to surfing or searching for accurate information are reported much less frequently than technological or social digital skills. Gender and age are significant in these two issues. These results underline the need for empowering children with human values and critical literacy, as part of a culture of digital rights and responsibilities. Derivado de la encuesta EU Kids Online (2018) este artículo analiza dos temas relacionados con la creciente difusión de mensajes de odio y noticias falsas: en primer lugar, cómo los niños y jóvenes italianos y portugueses (9-17) se comportan ante los contenidos y conductas negativas online. En segundo lugar, cómo aprenden a informarse y cómo gestionan las noticias. En comparación con las encuestas anteriores realizadas en Portugal e Italia (EU Kids Online 2010, Net Children Go Mobile, 2014), los resultados evidencian un aumento de experiencias negativas en línea (mensajes de odio, imágenes violentas y sangrientas, ser víctima o perpetrador de acoso cibernético). Aunque esto ocurre con más frecuencia en Portugal, los niños italianos y portugueses presentan patrones relativamente similares de afrontamiento. En ambos, la competencia navegando o buscando información precisa es menor que la competencia tecnológicas o sociales, afectando el género y la edad. Los resultados subrayan la necesidad de empoderar a los niños con valores humanistas y alfabetizaciones críticas, como parte de una cultura de derechos y responsabilidades digitales.


Introduction
The EU Kids Online network has been recognised as the primary source of high quality, independent and comprehensive evidence underpinning a better and safer internet for children in Europe.Involving 33 countries, the network integrates research expertise across multiple disciplines, combining methodological approaches, and has built constructive relationships with policy makers, media, industry, educators and practitioners at national, European and international levels.Its findings and reports are widely referred to in policy statements, having guided numerous initiatives to improve children's online experiences.The network sets out to continuously provide empirical evidence on children's and young people's online experiences and related opportunities and risks.This includes large-scale comparative surveys as well as smaller studies with focus on specific questions.
In 2010, the EU Kids Online survey on online risks and safety in 25 European countries included an openended question -what things on the internet would bother children around your age?The question preceded other questions on risks that were in the public agenda: pornography, sexting, meeting strangers and cyber bullying.The answers of more than 9.600 children (9-16 years old) regarding this open question were coded in accordance to the typology of risks developed by the network: content risks, the child as receiver of mass-produced online content; contact risks, the child as participant in an (adult-initiated) online contact; and conduct risks, the child as actor (perpetrator and/or victim) in a peer-to-peer exchange (Hasebrink, Livingstone & Haddon, 2008).
Results highlighted children's concern about a considerable diversity of online risks, the subtle nature of the circumstances that trigger such risks (from deliberate search to accidental exposition to) and the relevance of violent imagery (Livingstone, Kirwill, Ponte & Staksrud, 2014).Content related risks -pornography or sexual content (reported by 22%), violent content (18%) -and conduct-related risks including bullying (19%) were at the top.The importance given by children to violent content was noteworthy insofar as it received less attention than sexual content or bullying in awareness-raising initiatives (Livingstone et al., 2014).There were 235 explicit references to self-harm, suicide or anorexia/bulimia, 117 to racist content, 61 to hateful content, and 25 to ideological, religious or fundamentalist persuasion.Besides massive content, children were also upset by user-generated content (UGC), including content created and disseminated by peers in social networks.
Research agenda in this area needs a continuous update since the process of societal appropriation and integration of digital media changes both in terms of media access and supply content -fed by technical innovations, markets and industries -and in terms of societal and cultural practices -led by new cohorts of users, their societal and cultural backgrounds and communicative practices (Hasebrink, 2014).Thus, research must pay attention to children's changing online experiences in a longitudinal perspective (Hasebrink, 2014).Therefore, from 2010 onwards, the research on children and the internet has developed due to the design and implementation of new surveys on risks and opportunities, in Europe and elsewhere.
Net Children Go Mobile (NCGM), a sister project of EU Kids Online, was conducted in seven European countries (Belgium, Denmark, Ireland, Italy, Portugal, Romania and the UK) in 2013-2014.Combining national surveys (Mascheroni & Ólafsson, 2014) with interviews and focus groups with children, parents, teachers and youth professionals (Haddon & Vincent, 2014), NCGM explored whether the changing conditions of internet access and use introduced by mobile and convergent media led to greater, fewer or newer risks to children's online safety (Mascheroni & Cuman, 2014).
In Brazil, CETIC launched the TIC Kids Online survey based on the EU Kids Online framework in 2013 and the study has been conducted each year since then.Brazilian children remarkably reported being bothered by messages of racism and of class discrimination from peers.These results propelled a set of new questions about negative UGC, which were included in subsequent international surveys.Chile, Costa Rica, Peru and Uruguay have conducted similar research and with Brazil they form the Latin American Kids Online network.
Launched in 2015 and involving countries from all continents, Global Kids Online is a collaborative initiative held between the UNICEF Office of Research-Innocenti, the London School of Economics and Political Science (LSE), and the EU Kids Online network.The project links evidence with the on-going international dialogue in regard to policy and practical solutions for children's wellbeing and rights in the digital age, especially in the global South (Livingstone, Nandi, Banaji, & Stoilova, 2017).Global Kids Online developed a research toolkit that enables academics, governments, civil society and other actors to carry out reliable and standardised national research with children and their parents on the opportunities, risks and protective factors concerning children's internet use.Participant countries include Argentina, Ghana, South Africa, the Philippines, Serbia, Albania and Montenegro.
Since the research agenda moved from "how children engage with the internet as a medium to how they engage with the world mediated by the internet" (Livingstone, Mascheroni & Staksrud, 2018: 1103), the new pan-European survey aims to: 1) follow up on issues risen by technological innovations, including new digital devices, services and platforms, in order to provide empirical evidence on the contemporary media environment; 2) register new relevant social developments, such as the role of social media in the diffusion of political and religious extremism; 3) assure quality, coherence and comparability of measure indicators over time, in order to identify the main changes in children's online practices; and 4) ensure meaningful and reliable comparisons of children's online experiences at the European level, in order to identify similarities and differences, without disregarding local and contextual variations.Some of the first results were collected in two Latin countries, Italy and Portugal.Both participated in the EU Kids Online (2010) and Net Children Go Mobile (2013-2014).Furthermore, the country classification based on patterns of children's online opportunities, risks, harm, and parental mediation in 2010 (Helsper, Kalmus, Hasebrink, Sagvari & de Haan, 2013), placed Italy and Portugal in the Protected by Restrictions cohort of countries, the largest one which also included Mediterranean countries such as Spain, Greece and Turkey.In both countries, Italy and Portugal, restrictive parental mediation was at the top and children presented the lowest rate of risk and harm (2%; EU average value: 5%).These two reasons -common longitudinal data; the same low rates on risk and harm -justify attention to their most recent results.Furthermore, the article attempts to integrate the current challenges of the crescent dissemination of misinformation and hate speech.Not only it aims to shed light on the negative experiences that children and young people are facing online; it also looks at everyday practices of searching information and its implied skills.

Engaging with the world mediated by internet
The following part briefly recalls the permeability of the internet system -composed by search engines, news outlets, political blogs, video tutorials, tweets and social networks -to carrying all kind of content and how youth peer culture fits with social networking sites.
Having started as a closed, publicly funded and oriented network for professional communication, since the 1990s the internet has become a "deeply commercialized, increasingly banal space for the conduct of social life itself" (Couldry & Hepp, 2017: 80).Dimensions of the current process of digitalisationdiversification of devices, media convergence, ubiquity, fast pace of innovation and interconnected processes of datafication (Hepp & Hasebrink, 2018) -have impact on the way different generations of users are experiencing the media.
In regard to the dissemination of negative content, we will use Klein's (2017) framework of information laundering, based on the concepts of white propaganda (Jowett & O'Donnell, 1999) and technical ethos (Borrowman, 1999).Often provided by design, white propaganda goes unnoticed by the receiver and differs from grey and black propaganda, which go from selective facts to outright fabrications.While grey propaganda delves into false advertising and statistics' tampering, and black propaganda is typically the most recognised form of propaganda, "in today's media-savvy society, white propaganda might be the most pervasive of the three because of its effective ability to penetrate mainstream issues" (Klein, 2017: 27), acting in relation to the technological environment and culture.
The concept of techno-ethos underlines the credibility and authority constructed online through programming proficiency.For Klein (2017), this labour can be extremely effective with the Net Generation and its 'critical surfing', e.g., "the practice of giving credence to websites simply based on their professional designs, visual appeal, sophisticated options in media convergence and the fact they were found through trusted search engines" (Klein, 2017: 28).Thus, procedures for identifying and refusing radical ideas request not only critical informational literacy but also frames of human values.Furthermore, and according to the desensitization model, exposure to aggressive messages such as those found in videogames, movies, television or the internet may reduce automatic triggering of negative emotional reactions to images, words or thoughts of violence (Soral, Bilewicz & Winiewski, 2018).
Online social networks represent dynamic spaces where individual identities are expressed, formed and shared.Their activated central rational is an 'invitation to discourse' -that is, to engage, comment, link, share and circulate (Couldry, 2012).Young people are increasingly integrating many forms of social media into their intimate, social and political lives: "online risky content is interwoven into a youthful peer culture of sharing and daring as links are passed from child to child, discussed ritualistically the following day, and used in social judgments about group belonging or exclusion" (Livingstone et al., 2014).Some of these judgments and behaviours are expressions of hate.Hate speech, as defined by the Council of Europe, covers all forms of expressions, which spread, incite, promote or justify racial hatred, xenophobia, anti-Semitism or other forms of hatred based on intolerance, including cyber hate (Keen & Georgescu, 2016).As young people are active participants in the processes of producing and being exposed to hate speech online (Titley, 2014), in recent years the Council of Europe has addressed and battled hate speech by mobilising young people as actors and promoters of a culture of human rights and democratic citizenship, online and offline.
With the European landscape experiencing crescent radicalism and extremism, how are European children reporting online violent content and aggression?How are they facing the growing dissemination of online hate speech?How do they cope with unwanted online contacts and hate messages?What are their skills for dealing with the challenges of an over-saturated world of information and persuasion?Acknowledging children's and youth's cultures as well as the role they play as active agents is required in order to identify successful orientations grounded on digital rights and responsibilities.The longitudinal and comparative perspective of EU Kids Online research contributes to this knowledge.Results from Portugal and Italy provide the first answers.

Methodology
The The core questionnaire includes the following sections: a) Individual characteristics: socio-demographic, psychological and contextual variables, such as offline activities and risks; b) Internet uses: access, uses, technology, platforms, a couple of indicators that are constantly changing due to the diffusion of new devices; c) Online activities, balancing more the attention to risks with the attention to opportunities; d) Digital skills and digital literacy, integrating new approaches for measuring reported skills and tangible outcomes in terms of digital and social inclusion; e) Risks: bullying, pornography, sexting, meeting new people, negative UGC content and new risks related to practices and affordances (web reputation and privacy risks such as situations generated by the use of applications of self-tracking or by practices of 'sharenting', the online (over) exposure of children without their permission; f) Mediation: parental mediation, the role of schools and teachers, peer mediation and the local context.The EU Kids Online network also developed modules on emerging topics to be answered by 11-17 years old: Hate Speech; Internet of things/privacy; Digital Citizenship; Cyber bystanders; e-Health.
Despite sharing the same questionnaire, there are differences in the contexts in which each survey was applied that need to be addressed here.The Italian survey was funded through the Ministry of Education.The fieldwork, conducted in November-December 2017, followed the random walk method for finding and interviewing children in their households, the same approach used in the EU Kids Online 2010 and NCGM.A nationally representative sample of children (N=1001) answered directly to the interviewer, while sensitive questions related to risk and possible harmful experiences were self-reported in conditions of privacy.The Portuguese survey was funded through a private entity, DNS.PT, and received logistical support from the Directorate-General for Education.The fieldwork, conducted from March to June 2018, involved a representative sample of schools.Children from the 4th grade of basic education to the 12th grade (N= 1974) answered to the questionnaire in ICT rooms, in conditions of privacy.They had to read the questions and then reported their answers using CAPI system, a procedure that implied a greater expenditure of time and effort from the child.Italy added the modules on Hate Speech and Cyber bystanders, and Portugal added the modules on Digital Citizenship and Internet of Things.
The following section builds on the national reports of EU Kids Online II (

More upset, bothering and scared online situations
In 2018, the number of Italian and Portuguese children who reported having felt disturbed in the past year has more than doubled.The percentage raised from 6% to 13% in Italy and from 10% to 23% in Portugal.Figure 1 presents a longitudinal view of children's online negative experiences by age, gender and country.A gender gap is also noticeable in Portuguese interviewees over time.Girls present always higher rates than boys: in 2018, the percentage of girls who report having had negative experiences in the last year (27%) more than doubled in relation to 2014.In Italy, the rates remain stable across time and by gender.

How do children deal with these negative experiences?
Asking for support and help from others are considered among the most effective strategies for dealing with disturbing situations in order to reduce the emotional and psychological stress (Vandoninck, d'Haenens & Smahel, 2013).Blocking unwanted and harmful contacts, changing privacy settings, reporting the problem by clicking on an online 'report abuse' button, contacting an internet advisor or the Internet Service Provider are active individual strategies for dealing with those situations.Table 1 shows some similar results for Italian and Portuguese children when dealing with these situations.In both countries, a friend around their age is the first choice when looking for social support.This choice is followed by parents, thus evidencing the weight of affective ties and trust attached to peers and family.The quite similar percentage of children and young people -around one out of four -that have not talked to anyone may suggest that they feel able to cope with the situation alone, or that they are not sure they will receive support or they are afraid of the consequences.Interestingly, siblings are not among the main supporters.The secondary position may be due to their absence (for instance, 40% of Portuguese children do not live with siblings), to age differences or to the already identified relations of power and control in relation to younger siblings (Fortunati, 2011).Teachers and other significant adults share relatively low rates.
In 2014, the similar NCGM question was asked to children in the following way: If you were to experience something on the internet or when you were online in different devices that bothered you or made you feel upset, how likely or unlikely is it that you would talk with the following people?Mothers (71%), friends (57%) and fathers (54%) were the persons to whom the children were 'very' or 'rather' likely to turn to when they had any online experience that upset them.By contrast, the majority said it was 'very' or 'rather' unlikely that they would talk to teachers (64%), youth workers and other adults whose job is to help children.
Table 2 shows the results of individual reactions in relation to an upsetting situation.Although passive responses and behaviours -ignoring the problem, closing the window -are at the top in both countries, an active response -blocking the contact -co-leads in Portugal and occupies the third place in Italy.Changing privacy settings shares relatively low rates (10-12%), and reporting the problem online is residual among Italian users.Deleting messages from the other person suggests lack of digital literacy and may be associated to feeling guilty about what went wrong.
While longitudinal results evidence a growing number of children reporting having had bothering online experiences, Table 2 suggests a relatively high number of children who adopt a fatalist reaction and lack of digital skills on safety.In 2018, the number of confident children who are able to cope with online risks and know how to self-protect and assure their digital rights seems to be still relatively low.

Increasing exposure to inappropriate user-generated content (UGC)
UGC is used in the EU Kids Online research "to emphasise the often non-institutional, peer to peer nature of such material, permitting individuals or small groups to promote values, activities or knowledge that may be harmful for children" (Livingstone et al., 2011: 97).In comparison to previous results, the new survey indicates an increasing exposure to inappropriate content, with a prevalence of violent and hateful content.The exposition to and/or deliberate contact with negative content is particularly high in Portugal and registers a dramatic rise, as shown in Table 3. Italian interviewees firmly condemned terrorist acts (93%), the use of bombs (92%), the use of violence for fun (91%), for solving world problems (86%), and as a way of obtaining respect from others (85%).Three statements received lower full disagreement: the use of violence is acceptable when the person is insulted or made fun of because of his/her religion, origin or colour (72%); the use of violence for self-defence or if friends or family have been insulted (58%).Pre-adolescent and adolescent girls lead in the rates of full disapproval.
Hate speech was defined as the diffusion of online violent, discriminatory and offensive comments.Hate statements also present high rates of full disapproval when hate content is used for fun (95%) or for obtaining respect from others (93%).In certain circumstances, however, online hate messages or offensive comments receive less disagreement: 71% fully disagree with the statement that the use of hate speech is justified if someone has insulted your friends and family on the internet; the rates of full disagreement rise to 74% if the person was attacked first and to 78% in the case of having been insulted/teased because of his/her religion, origin and colour.
The experience of hate speech seems to be mainly an indirect exposure: only 3% said they had received offensive comments against themselves or their community.Most of the interviewees reported feeling sad, angry and full of hatred for the hate content they have seen and the younger age group (11)(12)(13) is also likely to express fear.Acting mainly as bystanders of these hate messages in the past year, 42% tried to help the victim while 58% did nothing about it.

The experience of cyber bullying
For a comprehensible definition of bullying -online and offline -the following text was introduced in the first questionnaire: As pointed by Livingstone et al. (2011), the group of children being bullied and the group of bullies may overlap, i.e. that some of those who bully others have also been bullied themselves.
In 2010, the results on the bullying experience placed at the top Estonian children: 43% reported having been bullied off and online, far above the average (14%).Portugal and Italy were at the bottom, with respectively 9% and 11%.In the NCGM survey, Italy (13%) and Portugal (10%) were again below the average (33%) while Romania (41%) was at the top.
Table 4 presents the NCGM and EU Kids Online II results of children that report having been bullied online and offline, and children who report having bullied others.While the Italian rates point to a relative stability, the Portuguese ones point to an impressive rise in the number of children reporting having been bullied online and offline.Furthermore, the percentage of Italian interviewees that report having bullied others in the last year (9%) is half of the Portuguese ones (17%).In Italy, the highest rate occurs among the aged 13-14 and in Portugal it is more reported by older adolescents.
Italian by-standers provide additional information: 19% report they have seen scenes of online bullying in the last year in social media and other platforms.Less than half report being highly (16%) or rather (25%) concerned with the victim, 54% report being a little bit concerned and 5% answer they are not concerned at all.The percentages rise with age: from 4% among the 9-10 to 24% among 13-14 and 22% among 15-17 years old, without relevant gender differences.About half of the bystanders (49%) of bully acts report they have tried to help the victim, while 50% report they did nothing and 1% report they encouraged the aggressor.Younger interviewees are more likely to do nothing (67%) and report more than the others they are not at all concerned with the victim.Gender also makes a difference: 60% of the girls reported they tried to help the victim, 61% of the boys reported they did nothing.

Informational and participatory practices and related skills
Feeling upset by something experienced on the internet, contacting with violent images, being exposed to hate messages and other negative UGC and cyber bullying point to the 'dark side' of the internet experience.Most young people identify and condemn violent content and behaviours, even if they do not act consequently.As noted in Table 2, the number of children that use the bottom of reporting abuse is quite low.
But what may be suggested in regard to their practices related to racism and discrimination that may be hidden in factual online sources?To what extent is informational online searching embedded in their everyday life?And how are children skilled to evaluate this kind of subtle content?To explore these questions, table 5 presents data on frequent use of the internet for informational and participatory purposes.The most frequent informational use of the internet is related to school work in Italy and in Portugal it reaches the second position.The main online participatory context is related to shared interests and hobbies.
Searching online news has increased in Portugal and presents a stable position in Italy.Looking for health information is rather reduced in both countries and discussing political and social problems is almost residual.Basically, this picture suggests that children and young people mainly use the informational dimension of the internet as students and to explore its participatory potential for a peer culture based on common interests.Bearing these trends in mind, it is worth seeing how skilled they report to be in relation to informational practices.
Following recent research on digital skills and related outcomes (van Deursen, Helsper & Eynon, 2014), the new questionnaire asked children on technological (T), informational (I), creative (C), social (S) and mobile (M) skills (Table 6).
Table 6: How children and young people position themselves concerning skills (2018) [7]   % The statement I find it easy to check if the information I find online is true receives the lowest percentage among the ten skills in Italy and the second lowest position in Portugal.More than half of the Italian children and half of the Portuguese ones do not consider easy to check whether the information they find online is true, thus suggesting they identify the challenge of this task in an overcharged informational environment.Reported by two out of three interviewees in both countries, being able to choose the best keywords for online searches generates fewer uncertainties.
Age and gender make a difference for informational skills.In Italy, among the 9-12 years old, only 28% of the boys and 13% of the girls agree with the statement that it is easy to check if the information is true.The level of self-confidence grows among adolescences and the gender gap is visible: 63% for boys, 53% for girls.In Portugal, girls also express less confidence than boys in considering that it is easy to check the veracity of online information (respectively 32% and 44% among the aged 9-12; 54% and 66% among the aged [13][14][15][16][17].
Agreement with the statement -it is easy to choose the best keywords for online searches -also rises with age in Italy and Portugal.It is expressed by around 60% among the aged 9-11 and by around 75% among adolescents.
The module on digital citizenship, answered by Portuguese interviewees aged 11+ (N= 611), adds information on youth's trust on news media and social media as sources of information, as shown in Table 7.
Table 7: Portuguese children's level of trust in the news (%) [8]

1) I think I can trust most news most of the time (N=626)
Fully agree/tend to agree 33 Fully disagree/tend to disagree 33

2) I think I can trust most news I choose to read (N=634)
Fully agree/tend to agree 38 Fully disagree/tend to disagree 28

3) The news media (television, radio, newspapers) do a good jog in helping me distinguish facts from fiction (N=608)
Fully agree/tend to agree 54 Fully disagree/tend to disagree 18 4) Social media (for instance, Facebook, Twitter, Snapchat, Instagram) do a good job in helping me distinguish facts from fiction (N=613) Fully agree/tend to agree 28 Fully disagree/tend to disagree 32 Source: Ponte & Baptista (2019) Trusting or not trusting most news most times receives similar percentages.However, 10% 'fully disagree' that news are mostly trustful against 3% that 'fully agree' with the statement.More interviewees agree than disagree with the statement in regard to trusting the news they choose to read.Here, the rates of 'full agreement' and 'full disagreement' are similar, respectively 9% and 8%.
Regarding the statement that informational mass media (television, radio, press) and social media help to distinguish between facts and fiction, more than half of the interviewees positively evaluate the role of informational media, twice more than social networks.The positive role of mass media for this distinction grows with age and girls express more this evaluation than boys.

Discussion
Recent results from Italy and Portugal demonstrate the challenges faced by children and young people in integrating the complexities of a digital landscape changing at a very fast pace.In 2018, disturbing situations, exposure to negative user-generated content -including violent images and hate speech -and cyber bullying rose in both countries and reached impressive rates in Portugal.This rise is particularly high not only among older adolescents but also among the youngest interviewees.In exploring more the online possibilities, they also face more riskier situations, in line with the identified positive correlation between opportunities and risks, the more... the more... (Livingstone et al., 2011).Moreover, these results contradict the ideas expressed by parents that their children are 'too young' for conversations about risky situations (Ponte & Simões, 2009) or that Portuguese and Italian children seemed to be 'protected by restrictions', as was noted in the first pan-European survey (Helsper et al., 2013).
Young people report that, by far, the sources of help they seek more often are a friend of the same age and their parents.The rates underline both of the relevance of peers at these ages and of the supportive role of the family across cultures.Nevertheless, the relatively high number of those that do not ask for help deserves attention as well.Do they feel they are sufficiently able to positively cope with the situation or are they afraid of possible restrictive consequences?While these rates are similar in Italy and Portugal -around one out of four children reported they have not talked to anyone about the situation -the differences suggest a higher rate of fatalistic behaviour and lower technological skills for managing unwanted online contacts among Italian children.However, the relatively low number of Italian children (130) that reported having had upsetting online experiences in the last year should be noted.
Besides content circulating in amateur videos, negative user-generated content include persuasive websites and blogs which present informational guidelines on ways of being very thin or on exploring physical and emotional limits, both situations that directly appeal to adolescents' peer culture and pressure.As noted by Klein (2017), the 'techno ethos' of those 'informational' platforms may contribute to the efficacy of 'white propaganda'.Such content also circulates in platforms where children and adolescents may participate, asking and discussing questions and looking at others' opinions.As seen, children enjoy a lot to participate in sites/forums where people share their interests.
Seeing gory and violent images exposes the nature of the circumstances, either accidentally while navigating or as a deliberate search for measuring emotional limits.In fact, research on viewing violent imagery has pointed to a lively peer culture of testing how much one can take or a curiosity about what the world contains outside the adult-imposed boundaries that constrain children, as noted by Livingstone et al (2014).
The rise of hate messages that attack certain groups or individuals is particularly noteworthy in Portugal, having been reported by two out of five interviewees.Although mainly sporadic, hate messages, cross all age groups.Furthermore, the intensity of their frequency (weekly, daily or almost daily) increases with age in adolescence, thus suggesting deliberate searches or being targeted for such content.In comparison to the Italian results and to 2014, the percentage of Portuguese interviewees that report cyber bullying has also dramatically increased, rising to one out of four children.
As far as the question of bystanders is concerned, it is remarkable that more than half of the Italian children that have seen hate messages or cyber bullying in the past year reported they have not reacted.As noted, the rates among the younger age group are particularly high, suggesting they do not feel able to have an intervention.
The relatively low rates reported in regard to informational skills suggest that there is a certain awareness of the challenges posed by searching for information.As pointed by Klein (2017), factual information sites are places where the connections to 'white propaganda' may be disseminated.At such sites, 'critical surfing' based on technological proficiency may replace the critical evaluation of sources and content.Furthermore, answers based on users' self-perceptions even suggest naivety, such as believing that it is easy to check whether the information found online is true.They may also believe that it is easy to choose the best key words for finding information.This evaluation is remarkable since around two out of five Portuguese interviewees aged 11 and older think they can trust most news they choose to read or use the internet for searching for information for school work.
Besides age, gender differences in dealing with disturbing situations and content and in dealing with information management also deserve attention.For instance, in questions about bystanders, Italian girls report far more often than boys that they tried to help the victim, suggesting they may be less exposed to the 'desensitization effect' (Soral et al., 2018).Furthermore, in both countries, girls seem to be less confident than boys in relation to their online skills, particularly skills related to navigating and searching for information.It may be said that besides the key factor of age, gender divisions on these issues pertain to the third level of the digital divide, the one related to self-perceptions of interest and opportunities achieved by using the online environment (van Deursen & Helsper, 2015).

Conclusions
Bearing in mind the complexities of "living the world through the internet" (Livingstone, Mascheroni & Staksrud, 2015: 1103), its technological affordances for disseminating all kinds of content as well as the particularities of youth peer culture, the focus on vulnerabilities has involved a longitudinal and comparative look on particular questions from the EU Kids Online survey.This inquiry does not ignore that the online experience of children and youth should not be reduced to negative situations.The internet includes opportunities for socialisation, having fun, participation and interaction, which also deserve consideration in the survey.
Comparing two countries allows us to place each one in a relative position.Their differences are quite marked in the data concerning disturbing situations and ways of dealing with them, negative UGC, cyber bullying, skills and search for information.In regard to other topics, such as the key people to talk to when something negative happens, the results are quite similar.One should not forget that children's answers are not only based on self-reporting -whose rates may not coincide with the ones concerning outcomes -but also that the surveys were given under differing conditions.To what extent the Portuguese results were affected by the fact that data collection occurred at schools is a question that must wait for further results from other countries.Furthermore, the broader picture provided by these quantitative results should be addressed with qualitative approaches (e.g.observation and discussion of practices and situations).Positioning children and young people as active discussants of these results is part of a next step, following the activation of a participatory practice of the Global Kids Online.
A final note addresses hate speech as one of the most harmful effects of the internet system.Governments, social platforms and news media can certainly contribute towards reducing hate speech, but all the institutions that are closer to young people -from family to school and local communities -also have roles.Moreover, young people should be recognised not only as vulnerable victims but also as active agents against such messages of cyber hate, as the recent Council of Europe campaign pointed out.In line with this awareness raising, the final words belong to answers given by Portuguese adolescents on what bothers people their age: "Seeing people mocking at refugees."(Boy, 13 years) "Disagreement of opinions, discrimination from other people and the way they look at differences."(Girl, 15 years) "What bothers me the most is the fact that people use the internet to offend or denigrate someone else's image.Bullying because the person is black, ugly or because of their sexual option."(Boy, 16 years).
"Comments or photos of a prejudiced nature.Dissemination of terrorist attacks as something positive.Negative comments regarding a group with a different religion, sexual orientation, among others."(Girl, 17 years).

Bibliographical references
article presents Italian and Portuguese results (Mascheroni & Ólafsson, 2018; Ponte & Baptista, 2019) based on the core EU Kids Online questionnaire and optional modules.Children's age range is now extended to 17 years old, in line with the UN Convention of Children's Rights.

Figure 1 -
Figure 1 -Online negative experiences, by age and gender (2010, 2014, 2018) [1] Italy Portugal of those who answer 'somewhat true of me' and 'very true of me' Italy Portugal T -I know how to save a photo that I find online 80 79 T -I know how to change my privacy settings (e.g. on a social network site) 65 81 I -I find it easy to check if the information I find online is true 42 52 I -I find it easy to choose the best keywords for online searches 68 66 S -I know which information I should and shouldn't share online 78 88 S -I know how to remove people from my contact list 80 89 C -I know how to create and post online video or music 66 62 C -I know how to edit or make basic changes to online content that others have created 48 37 M -I know how to install apps on a mobile device (e.g.phone or tablet) 85 92 M -I know how to keep track of the costs of mobile app-use 56 66 Sources: Mascheroni & Ólafsson (2018); Ponte & Baptista (2019)

Table 3 : Children who had seen or discussed negative user-generated content (2014-2018) [4] % of those who had seen or discussed…o… NCGM Italy
Having seen gory or violent images -a new item -leads in both countries, and the others evidence a remarkable rise from 2014 to 2018.In Portugal, the rise of hate messages and ways of physically harm or hurt themselves is particularly noteworthy.Certainly, the nature of the circumstances that usher such experiences (from deliberate search to accidental exposition to) needs to be considered.Figure2presents the rates of hate messages attacking groups or individuals and the rates on gory and violent images by age and gender regarding Portuguese children aged 11 and older (N=1280).

Hate messages and gory and violent images, by age and gender (Portugal, 2018) Hate messages attacking groups or individuals Gory and violent images
16% of the adolescents aged15-17.Hate speech and other negative content seem to be not only reaching adolescents but also younger ages.The specific module on violence and cyber hate answered by Italian interviewees aged 11 and older (N= 707) provides more information on children's reactions to violent acts and hate speech.They were invited to position themselves in a five points scale concerning statements on violence: 1) it is acceptable to use violence against others for self-defence; 2) it is acceptable to use violence against anyone who insulted