Two-thirds of Australian children surveyed report grooming-related contact from online strangers, Adelaide University study finds

Earlier prevalence studies asked a narrower question. They generally counted overtly sexual or unwanted solicitation from someone the child believed to be an adult, and they returned rates from about 5% to 30% ^[13][14]. Katie Logos and her co-authors counted the approach behaviours as well, including questions about school and whether parents check the phone, and asked about any online stranger rather than only suspected adults ^[14][15][16]. On that wider construct, nearly two-thirds works out at roughly 2.2 times the top of the old range ^[2], which is close to what the definition change alone would predict. Whether Australian children are approached more often than they were five years ago is not a question this design can answer.

Denominator discipline matters here. Two-thirds of about 300 children is roughly 200 respondents; the third reporting direct sexual requests is about 100, and the one in five asked for photos or a meeting is about 60 ^[1]. Had the sample been drawn at random, the 95% interval around two-thirds would run from about 61% to 71% ^[3]. It was not drawn that way: the authors built it to broadly reflect the demographic characteristics of the Australian population ^[4]. Because the sample was not drawn at random, that five-point margin marks only the floor of the real uncertainty.

The finding most open to misuse is the null one. Children who used privacy settings, parental guidance or age-appropriate platforms were just as likely to report grooming-related communication as those who did not ^[10]. Professor Russell Brewer, who directs the Australian Cybercrime Observatory, points out that the arrow may run backwards, with some children adopting stronger practices after such an interaction ^[11]. One measurement at one time cannot separate a safeguard that failed from a safeguard adopted in response ^[11].

What the survey cannot say is how many of those strangers were adults. Children often believed they were talking to another child ^[8], and Logos notes that they cannot verify who is behind an account, which is where the concern about adults posing as children sits ^[18]. This means the survey measures grooming-related contacts, and the number of distinct offenders behind those contacts remains unknown. The behavioural part is the part that holds: the opening moves arrive inside conversations that read as friendly ^[17], and the common responses in this sample were to do nothing or to block the sender ^[8]. Detection systems that rely on a child reporting distress therefore face a gap: most children in this survey did not report feeling that distress at all.