statistical power of a General Linear Mixed Model with repeated measurements

I would like to be able to calculate the statistical power of a General Linear Mixed Model.
This model is applied on longitudinal data on dives of cormorants. I have for the analysis 2 groups of 6 cormorants each. The goal of this study is to test the effects of a localisation device (GLS) on the dive abilities of these birds. Each bird was carrying a logger in order to record some dive variable (depth, dive duration, descending and ascending rates...).

The number of individual is pretty low but the number of observations (dives) per bird is quite high (mean 416,5 in one group and 407 in the other).

We compared dive parameters between the two groups of birds with General Linear Mixed Models where bird identity was included as a random factor (in order to take the correlation among repeated measurements into consideration). As all variables are dependent on the depth of the dive, we included dive depth as a covariate

I tried to use the procedure GLMPower in SAS but apparently it does not allow to consider a random effect. Moreover this procedure allows only Prospectives analysis (to fix the number of individual in order to obtain some power and so to do before the data collection) but not retrospectives analysis (where someone wants to achieve the statistical power once the study have been done).

I am looking for a way to calculate such a power. I can use SAS, SPSS and R. I can also use any available freeware to do it, but I don't know which one can do such calculations for a General Linear Mixed Model.

Last edited: