The 95% confidence interval is 5g - 40g, which seems pretty wide to me. I would like to know at what confidence threshold 0g lies. It seems to me that there is a good chance this result is just random chance. Am I missing something?
Based on the normal distribution, 0 should be somewhere around 97% confidence. But that becomes suddenly less impressive when you consider the fact that they looked at each month separately. So after looking at 9 numbers, they found 1 that was 1/30. The odds of seeing THAT are over 30%!
So yes, this does seem like a chance result to me. As the saying goes, "You torture the data until it confesses."
The confod
and that's how the revolution starts, those heavy babies grow up to overthrow the communist government.