> Interesting. I'd be curious to see how the hours spent at work impact that number. I'de more interested in comparing these percent changes with the sample's

> Interesting. I'd be curious to see how the hours spent at work impact that number. I'de more interested in comparing these percent changes with the sample's margin of error. Extrapolating conclusions based on a deviation of 0.8% sounds an awful lot like grasping at straws.