This is a G*Power question:

The assignment our statistics teacher has given us is...

What is the difference in Barthel scores over time?

I have 3 data sets- Barthel 1, Barthel 2, Barthel 3. Each of these data sets are the results of post-stroke assessment scores in the Barthel post stroke at 1 month, 2 month and 3-month intervals.

I have to first use the G*Power software to determine the appropriate sample number for the 1000 participant data set given.

Here is the problem I am having...What do I select as the test family and statistical test and the numerator portion?

In G*Power, I'm thinking I choose: "f-tests" for the test family, then "ANOVA: Fixed effects, special, main effects and interactions" for the statistical test. Effect size: 0.3, alpha err prob: 0.05, Power (1-beta prob): 0.90, numerator df;???, number of groups: 3.

Can you help me understand which ANOVA I am to choose and why is it the choice as well as the numerator df?