Quote:
Originally Posted by Alex
Yup! Both of those points may affect the results a bit. You can certainly normalize the groups as much as possible by trying to match years of riding experience, type of riding, miles per year, and perhaps a few more variables to try and make a little more sense of the data, but doing all that may still leave some leanings in the data that can be interpreted a few different ways.
|
Agreed. Get one of those statsticians to really get the proper variates and pull a good standard deviation to see what conclusions we can really draw.
My (developed) instinct is to always questions statistics. Most stats and science is incorrect. In other words, given more time and new information, the original conclusions of the observed data will be proved false.
Those who buy safety oriented stuff tend to be...uh...safer. So, attribution gets very tricky...