Type 1 2 Errors
- May 31, 2007 at 1:09 pm #47128
If I run a proportion or average hypothesis test and obtain a p-value less than .05, do I need to complete a power and sample size analysis to assess the probability of detecting a certain shift with certain sample size?0May 31, 2007 at 4:20 pm #156782
annonParticipant@annon Include @annon in your post and this person will
be notified via email.
You can run the Power and Sample Size test via MTB before or after data collection, depending on where you are in your project.
Ideally, you run it before so as to collect the proper sample size for a given level of precision and power. If you have already collected your data, then at least run it so as to quantify the above elements given your current sample size.
Always consider what questions your data needs to answer before you get to collection…this allows you to choose the proper data form and subsequent analytics….I would rather design the data collection around the analytics than the analytics around the data collection…just my opinion.
The forum ‘General’ is closed to new topics and replies.