It's 18-24 year-old people (or people claiming to be 18-24 years old) answering survey questions under who knows what circumstances. A large fraction is likely

It's 18-24 year-old people (or people claiming to be 18-24 years old) answering survey questions under who knows what circumstances. A large fraction is likely intentionally trolling the pollsters to rile up oldsters, and some might be more like 13 and don't quite understand the questions. I think the funniest example is this pair of questions: There was an explosion at a Gaza hospital. Do you think the explosion was caused by an Israeli airstrike or by a terrorist rocket that went off course? 45% say "Caused by Israeli airstrike", 55% say "Terrorist rocket that went off course" But then... Given that US intelligence and Israeli intelligence have said and shown evidence that the hospital explosion was caused by an off-course terrorist rocket do you think the explosion was caused by an Israeli airstrike or by a terrorist rocket that went off 61% say "Caused by Israeli airstrike", 39% say "Terrorist rocket that went off course" Mentioning the involvement of intelligence services makes people less likely to agree with their explanation! (This seems to affect age groups up to 44 to some degree.)