Hello Bhoomin B Chauhan. You say you have 3 DVs, participant workload, situation awareness, and performance. If you use MANOVA, your DV is in essence an optimal (by some definition) linear combination of those 3 DVs. So my question to you is this: Do you really have one multivariate question that uses that linear combination as the (effective) DV? Or do you have 3 univariate research questions, one per DV? I would be surprised if it is not the latter situation. For more discussion of these issues, see the following articles and blog post. HTH.

Article MANOVA: A Procedure Whose Time Has Passed?

Chapter Multivariate analysis versus multiple univariate analyses

http://psychologicalstatistics.blogspot.com/2021/08/i-will-not-ever-never-run-manova.html

To perform a power analysis for a within-subjects MANOVA in G*Power, you would need to select the "Repeated measures, within-between interaction" option under "Test family" and "Multivariate: Repeated measures, within-between interaction" under "Statistical test."

For the input parameters, you would need to specify the following:

• "Number of groups": Since you have two IVs, each with two levels, you have four conditions in total, so you would enter "4" as the number of groups.
• "Number of measurements": You have three DVs, so you would enter "3" as the number of measurements.
• "Correlation among rep measures": This refers to the correlation among the DVs across time points. You can estimate this based on previous research or pilot data. For the purposes of the power analysis, you can assume a conservative estimate of 0.5.
• "Effect size f^2": This is the effect size for the MANOVA. You can estimate this based on previous research or pilot data. A commonly used rule of thumb is that a medium effect size corresponds to an f^2 of 0.15. You could start with this value and adjust it based on your specific hypotheses.
• "Alpha error probability": This is the significance level for the test. You can set this to 0.05, which is the conventional value.
• "Power (1-beta error probability)": This is the desired level of statistical power. A common value for statistical power is 0.80.

After entering these input parameters, you can click "Calculate" to determine the minimum sample size needed to achieve the desired level of statistical power.

In terms of selecting the appropriate test, "Repeated measures, within-between interaction" is the correct option to use for a within-subjects MANOVA with two IVs and three DVs. This option will allow you to test for the interaction between the two IVs, as specified in your research hypotheses.