That's an interesting study paper. I think they're making a mistake in wording the conclusion with so much certainty based on the tiny sample size though. I'd

That's an interesting study paper. I think they're making a mistake in wording the conclusion with so much certainty based on the tiny sample size though. I'd say big companies that can confidently stratify the age of users in their porn search preferences, or even social media users, would be a good way to get a more valid sample size.