Originally Posted by
Wesley36
Alright, if you want to be pedantic, 145 cranks did have higher output, but according to statistical tests, it is not clear that the difference is significant (ie that it was not the product of random variation). It is true that the null hypothesis (there is no difference between 145 and 170) is not disconfirmed,
but it is not true that the tests showed no difference.
No, it means exactly that. Those power numbers are not different, and saying they are is wishful thinking.
Originally Posted by
Wesley36
To say the difference is not statistically significant does not mean they are the same, it means that the difference could be random chance. Could be random chance does not mean is random chance. But you are right, the only clear conclusion is that 145mm and 170mm cranks produce more power than 120 and 220mm cranks.
The appropriate statistical tests tell you whether the two sets of numbers are the same, or are not the same. In this came, they are the same.
If you can't prove the study's hypothesis, you HAVE to accept the null hypothesis (that there's no difference). You can't say "well, there is no difference in my study, but I want to believe there is, so I'm not going to accept these results and will continue to believe what I want, in spite of what the numbers say." Well you
can say that, but then you'd be wrong.
Bottom line, if you went by these results, you should be able to generate the same power on cranks that range from 145 to 170mm in length. Not a great advertisement for the PwerCrank product, if that's what they're selling.