Talk:Simpson's paradox

More.... this seems to not tell a story... yet? ħuman 04:07, 21 September 2009 (UTC)

This has been on my to do list for ages, it is actually quite important as it is often exploited by woo-miesters and dodgy pollsters. Expect exciting developments in the next 15 to 30 days :) - π 04:17, 21 September 2009 (UTC)

Operation is expected soon; needs awesome title like "Pie for apples" ħuman 04:20, 21 September 2009 (UTC)

How about another example to further demonstrate: You have two boxes, each with 6- and 20-sided dice. In one box, there are 8 6-sided dice and 2 20-sided dice, and each die has one face marked. In the other, there are 2 6-sided dice and 8 20-sided dice, and each die has 2 faces marked. Each individual die from box 2 is twice as likely to land marked face up as the corresponding dice from side 1, but since there are more 6-sided dice in box 1, a randomly selected die from box 1 is more likely to land marked face up than a randomly selected die from box 2. 166.137.8.88 (talk) 22:57, 30 September 2014 (UTC)

Please don't mention 20-sided dice, I'll have Warhammer flashbacks. Sophie^Wilder 09:05, 1 October 2014 (UTC)