Hi Experts,
I am trying to choose my inference approach: fixed vs mixed effects models, for my ANOVAs. I am also a bit confused as to whether my factors are fixed or mixed, and would be grateful for some advice...
Factor A is eating rate condition: normal rate or slow rate (10 participants were randomly assigned to each condition, therefore only 20 in total).
Factor B is 3 levels on a psychological task.
I am also interested in the interaction.
I think they are both fixed (only two eating rates of interest and only 3 levels on the task). But I am concerned Factor A may in fact in a random effect...
As an aside, I have done LOTS of reading on here and decided I needed to average the 3 levels on the task per subjects at 2nd level and run a group difference at the 3rd level for the main effect of group. Then run a 2 way mixed effects ANOVA, as per the GLM page for the main effect of task and interaction with group. Hope this sounds sensible...
Then I have ran my ANOVAs, but not certain whether these should be fixed effect or mixed effect models in the Stats tab. I understand that I can only make inferences about these exact participants if using fixed effects but perhaps with the relatively low number of subjects this is reasonable for this study? Or should I use a mixed effects model then I can make inferences about the population (albeit simple OLS as I understand I can't use FLAME without the varcopes...).
Some clarification is much needed, so many thanks in advance for your help!
best wishes
Ella
|