Hello,
I have a dataset of an objective color measure (color harmony, a numerical value) of images of 24 different objects. Each object was captured under 8 different light sources via 3 different cameras.
What I want to do is study the impact of the camera (3), light source (8) and the choice of object (24) on the color harmony scores (576 scores), both individual impact and interactions between them.
Since my color harmony data is not normally distributed, I have to use non-parametric tests.
I tried using Aligned-rank treatment but I get the error dF=0 (which is normal as I have total 576 scores=3x8x24). I don't have any sort of repetition in data. I can't use a Generalized Linear model either as I do not have any random effect grouping factor.
Can somebody please help me in choosing the right statistical test and a follow-up post-hoc test as well?
A big thanks in advance!