I have data on continuous measurements (length of time of a behaviour) from two groups of individuals which differ in their phenotype (phenotype A or B, the variable used to group them). Each group is made up of a small number of individuals (4-5), and I have multiple (>100) measurements from every individual within each group. The number of individuals per group and the number of measurements per individual are not equal.
How would I go about statistically comparing the mean and variances of these measurements between groups (phenotypes), accounting for the fact that each measurement is not independent (i.e. there are multiple observations from each individual within the groups, rather than each measurement being from a different individual? I basically want to know if the length of time spent in a particular behaviour differs between phenotype A and phenotype B - do the means differ and is one phenotype more consistent (have a tighter distribution) than the other.
Many thanks in advance!