no matter how you do your analysis the decision will be the same? That is, if you do your test based on: 1) An acceptance range built around the population or hypothesised value. 2) A confidence interval around the point estimate (e.g., mean). 3) A calculated t statistc compared to a critical t value (e.g., if t_calc > t_critical then reject). I could check this with an example, which will take ome time, but would like to hear from you whether this is true or not. This should also work whether you have one or two tail tests, or at least I’m claiming that.
Anyone wants to say yes or no?
"no matter how you do your analysis the decision will be the same? " Schweser seems to find a way to do it wrong all the time. But each of these methods will work.
Looks logical!