3-way Anova_ Lost DFs problem

[#1]

Hello,

in my experiment I have two group of different cultures, each group is made of female and male (gender) and each participants can give correct and incorrect response, to the stimuli which are black/n/white or coloured. I tryed to perfom a 3-way anova with the fit model Y=response time (continuos variable) and X I did a full cross with culture, gender and colour; I also nest subjects with gender. In some of the effects the outcome shows Lost DFs, what this could be due to?

Thank you:)

[#2]

Generally that type of message from JMP is a signal that you are attempting to estimate effects for which there is not sufficient combinations in the experimental space for their estimation. If you'd like us to try and see the root cause, if you could include your experimental design and model that you are trying to estimate we might be able to offer a solution.