Thanks! :)

I have one final set of questions regarding this particular design (as described in the first post). We do have 20 subjects and all of these 20 are associated with both levels of the DRUG factor (they all got both Drug and one Placebo). But only 10 of them are associated with the each level of the GROUP factor (which background the drug was paired with). So, should -clevels be 10 or 20 when we do the nested -type 5? I feel it would be 10 and then you have to make sure that Factor A is really the one that C is nested in.


Also, with this design. Can't you also do somehing like Factor 1 = Drug or Placebo, Factor 2 = What background are they looking at (#1 or #2)? Factor 3 = Subject. So we would get:

1 2 1 (Looking at drug Picture, Picutre is motive #2, sujbect 1)
2 1 1 (Looking at placebo Picture, Picutre is motive #1, sujbect 1)

1 1 1 (Looking at drug Picture, Picutre is motive #1, sujbect 2)
2 2 1 (Looking at placebo Picture, Picutre is motive #2, sujbect 2)
...

1 2 20 (Looking at drug Picture, Picutre is motive #2, sujbect 20)
2 1 20 (Looking at placebo Picture, Picutre is motive #1, sujbect 20)


1. Would this be OK?
2. What would the difference be, compared to the way we did it above with type 5?
3. In this design all subject belong to both levels of borth groups so I guess this would be a type 4?



Edited 2 time(s). Last edit at 04/20/2016 07:01AM by Robin.