1
$\begingroup$

I have made an experiment with 28 subjects. All made the experiments under 3 different conditions (A, B and C). So I have a table like this:

No. A B C
1   3 6 4
2   6 2 6
...

Now I just want to perform a RM-ANOVA or a Friedman test or a Quade test. That's not the problem.

I want to use GPower to determine the necessary alpha value to detect e. g. a "medium" effect. I do not get how to use GPower in this case.

Here is a screenshot of how I do it for a t-test:

enter image description here

Now I want to do the same for more than 2 conditions. Can somebody point out how to do that? Here's a screenshot of the possibilities:

enter image description here

EDIT: I looked it up, and apparently it is called One-way repeated measures ANOVA. So how can I to One-way repeated measures ANOVA with G*Power to determine the required alpha value (given effect size and so on)?

$\endgroup$

1 Answer 1

2
$\begingroup$

I think I found it here:

Link to book chapter, additional material

Applied to my problem I guess it will be like this:

enter image description here

F tests - ANOVA: Repeated measures, within factors
Analysis:   Criterion: Compute required α 
Input:  Effect size f   =   0.25
    Power (1-β err prob)    =   0.80
    Total sample size   =   28
    Number of groups    =   1
    Number of measurements  =   3
    Corr among rep measures =   0.3
    Nonsphericity correction ε  =   1
Output: Noncentrality parameter λ   =   7.5000000
    Critical F  =   2.2129741
    Numerator df    =   2.0000000
    Denominator df  =   54.0000000
    α err prob  =   0.1191999
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.