Reply to thread

Message: <blockquote data-quote="orsal" data-source="post: 1353733" data-attributes="member: 16016">[spoiler]The difference: the scenario you describe in AU2 would be twice as likely to happen with a mother of two boys, since either of the boys might have been the one to run up.[/spoiler]Here's a similar, slightly easier to grasp, paradox: you want to get information about average family size. You conduct two surveys, one of a random sample of parents , the other of a random sample of children. You ask the parents how many children they have, and compute the mean average. You also ask the children how many children in their family, and compute the mean average. You get a significantly higher average family size from the second sample. Why? [spoiler]Because the larger the family, the more likely it is to get into your sample. A family with 6 kids is six times as likely to get sampled than a family with only one kid. But both are equally likely to make it into your survey of parents.[/spoiler]</blockquote>