Hi, I have a complicated statistical problem that I wonder whether I can get some advice on here:
I have an uneven number of groups in a GLM model. There are two groups of male and female mice tested for several dependent variables under two conditions: experience and no experience; each of these groups (experienced and inexperienced male and females) have been further tested under one more condition (introducing reinforcements). The issue arises when the males have been tested with both positive and negative reinforcements, but females only with positive. Now when designing the GLM model, I can look at the main effect of: sex, reinforcement, experience, and the interactions of sex*experience, reinforcement*experience but not the 3-way interaction due to lack of females in the negative reinforcement group. The post-hoc results, thus fails to give me a clear result for males in positive reinforcement interaction group due to either being grouped with males (sex*experience interaction) or in positive reinforcement with females (reinforcement*experience). Could you advise if there is a way to resolve this considering I cannot do further experiments to test females in negative reinforcement situation?
- Badges
- Science topic
- Similar topics
- Psychology