(Step-by-Step) OK, you now know the seven steps for hypothesis testing. Step 3 made a big deal of setting the level of significance (or its supplement, the
Question:
OK, you now know the seven steps for hypothesis testing. Step 3 made a big deal of setting the level of significance (or its supplement, the confidence interval). Why is this such a big deal? What are we saying when we set the alpha (the LOS) higher or lower in value? Why would we want to use, say, .01 rather than .05 as a LOS for a hypothesis test?
First, setting LOS is a "big deal" because the level we set it at tells us how RISK TOLERENT we are to making a Type I Error. In other words, if we can tolerate making a Type I Error, we can set LOS to a higher level than otherwise. If we can't tolerate making a Type I Error, we have to set alpha (the LOS) to a lower value, say, .01 or even .005.
Second, why should we tolerate the possibility of a Type I Error at all? Why not just set alpha = 0 and completely remove the possibility of rejecting a true null hypo? LET's HEAR YOUR OPIONIONs on WHY WE CAN'T SET ALPHA = 0...or maybe we can.
Deliverable: Word Document 