How to determine % of people who keep secrets. Help frame search criteria.

#1
Sorry if this is the wrong place to post this question but I am having a very hard time searching for an answer about statistics.

I'm an 8th grade math teacher and I had a student ask me a question that I was not able to answer and I am hoping for some help finding the answer.

So, we were doing a presentation on internet safety and the video we were watching had some statistics mentioned. I'll paraphrase here and the numbers are made up.

The video said, "68% of all teen boys have an unsafe encounter with a stranger online." It then said that "32% of these teens never tell anyone about it."

My student then asked, "If they never tell anyone about the encounter how do we know come up with that 32% value."

My first response was through anonymous surveys but I've had enough statistics to know that this type of information can be inferred from actual data. However I don't remember how this is done.

I spent about 30 minutes on Google trying to search for an answer but I could not get anything that answered the question. Can someone help me with name of the methodology that is used or more specific search terms that will allow me to answer this question.

Again, sorry for the post in the wrong place. I really don't even know which subform would be best.

Thanks for all your help!

mtolbert@usd259,net
 

spunky

Smelly poop man with doo doo pants.
#2
My gut feeling is that your video simply misrepresented the statement "32% of these teens never tell anyone about it". By "anyone" they probably meant someone like a parent, a family member, a teacher or someone of relevance to the boy's life. I'm also thinking this could be a survey done on college students or young adults asking them about their online behaviour as teenagers. If they never told anyone (until the moment they answered the survey), then technically they never told anyone while being teenagers. This information only came to light once they became adults.

To be honest with you, I do think the first option (i.e. lack of clarification of what "anyone" means) is probably what happened in that video. Surveys like this don't usually go into complex statistical modeling and their primary function is mostly as descriptive devices.