Calculating sample size for a two way MANOVA using gpower?

#1
Hi!

Just wondering if anyone can help me with this?

My normal go to textbook - Quantitative Psychology by Clark-Carter - doesn't cover multivariate stats in great depth and for my project I am planning to do a two way MANOVA. The free Gpower 3 has been recommended to me, but there are no instructions, as such, on how to use it.

My design is as follows:

IV's

Gender (2 levels - male/female)
Age (3 Levels - Y7, Y11 and Students under 25)

DV's

Depression score
Aggression Score
PARQ mother score
PARQ father score
PARQ mean participant score
Attachment score

If anyone could help me with the following queries who knows how to use gpower, I'd be really grateful!

1. Which do I select for my design out of the following?

MANOVA: Global effects
MANOVA: Special effects and interactions
MANOVA: Repeated measures, between factors
MANOVA: Repeated measures, within factors
MANOVA: Repeated measures, within-between interaction

I want my alpha set and 0.05 and power of 0.8 but am unsure on the rest of the boxes required, particularly effect size f.

2. How do I calculate/find effect size f? I would like at least a medium effect size. so what value do I put here?

3. Does participant mean count as a measurement here, as it is not a measurement I am taking directly from the participants?

4. What is 'corr among repeated measures?



Sorry if this is coming across as silly to those in the know but I'm getting so lost!


Thanks in advance!



Becky
 
#2
Hi, it’s been along time that you ask the question but I’ll try anyway. I’m struggling with the similar question as well. Could you help me? Thanks in advance.