I have seen instances recently where statisticians are performing retrospective power analysis to determine the minimum detectable effect size for each parameter measured. The model is a repeated measures ANCOVA with difference at baseline as the covariate. There are 2 groups, n=8 per group, with 10 timepoints within 24 hours. For the methods of retrospective power analysis, they estimate variability using error degrees freedom and the standard error of the difference between control and treated groups in the model. Then the minimum detectable change from the control group is calculated based on this estimate of variability. Can anyone point me in the right direction or give a brief summary of how to caluclate the SD estimate and then the minimum detectable change?