1) Use G*Power to calculate sample size, but this is sometimes a little bit limited and assumes that all assumptions are met

2) Simulate your design, for example in R, and vary your important parameters and assess power over many samples.

Rainer Duesing

Thank you! I am going to use G*Power.

Dear Emily,

The following link is helpful. Please refer to method section of this manuscript.

https://www.researchgate.net/publication/339553439_Quantifying_lower_limb_inter-joint_coordination_and_coordination_variability_after_four-month_wearing_arch_support_foot_orthoses_in_children_with_flexible_flat_feet