Yeah but then you get headlines like 10% of children get cancer. How many children? 10 children? A million children? When the sample sizes are small, it's bett

Yeah but then you get headlines like 10% of children get cancer. How many children? 10 children? A million children? When the sample sizes are small, it's better to just give the fraction IMO.