How safe do youngsters really behave on Facebook: an observation-study

We are witnessing the rapid growth of a new generation of participatory and collaborative network technologies that provide individuals with a platform for sophisticated online social interaction. Social networks today have hundreds of millions of users and are transforming our social and professional interactions and to some extent are shaping society itself. This causes a growing concern about security and privacy issues in social networks, particularly with youngsters (Debatin et al., 2009; Jones & Soltren, 2005). To know the actual risks, we need to map out the state of the art with regard to youngsters and their behavior on social networks. A lot of survey-research has been done with regard to the amount and type of personal information children actually provide on their social network profile (eg. EU Kids Online survey). Yet, survey-research has its disadvantages: people may answer incorrect because they think it’s more appropriate or because they just don’t know the right answer. Especially in the case of sharing information on a social network-profile, the latter may frequently be the case. Some kids think for example that nobody can see the information they put on their profile, while actually everybody is able to see it. Therefore , this kind of research should go further: we tried to verify and extend survey-results by observing a thousand Facebook-profiles of Flemish youngsters aged 13 to 18. We selected the profiles as randomly as possible, but controlled for age, education and sex. We observed ‘friends’-pages and ‘friends-of-friends’ pages, to know to what extent youngsters change their privacysettings. Furthermore, the kind of information displayed on the profiles has been analyzed in terms of riskiness, for example, risky pictures, signs of bad behavior, drug abuse, alcohol abuse, aggressive behavior and so on. This riskiness was compared between groups of different age, education and sex. The found results are compared to the results of survey-research and interpreted in terms of risks. We conclude with some suggestions for education.We are witnessing the rapid growth of a new generation of participatory and collaborative network technologies that provide individuals with a platform for sophisticated online social interaction. Social networks today have hundreds of millions of users and are transforming our social and professional interactions and to some extent are shaping society itself. This causes a growing concern about security and privacy issues in social networks, particularly with youngsters (Debatin et al., 2009; Jones & Soltren, 2005). To know the actual risks, we need to map out the state of the art with regard to youngsters and their behavior on social networks. A lot of survey-research has been done with regard to the amount and type of personal information children actually provide on their social network profile (eg. EU Kids Online survey). Yet, survey-research has its disadvantages: people may answer incorrect because they think it’s more appropriate or because they just don’t know the right answer. Especially in the case of sharing information on a social network-profile, the latter may frequently be the case. Some kids think for example that nobody can see the information they put on their profile, while actually everybody is able to see it. Therefore , this kind of research should go further: we tried to verify and extend survey-results by observing a thousand Facebook-profiles of Flemish youngsters aged 13 to 18. We selected the profiles as randomly as possible, but controlled for age, education and sex. We observed ‘friends’-pages and ‘friends-of-friends’ pages, to know to what extent youngsters change their privacysettings. Furthermore, the kind of information displayed on the profiles has been analyzed in terms of riskiness, for example, risky pictures, signs of bad behavior, drug abuse, alcohol abuse, aggressive behavior and so on. This riskiness was compared between groups of different age, education and sex. The found results are compared to the results of survey-research and interpreted in terms of risks. We conclude with some suggestions for education.C