Good day scholars,
I have a paradigm divided into two main conditions. A baseline condition and an experimental condition.
In the baseline condition, participants answered to 20 identical trials.
In the experimental condition, participants answered to 50 (modified) identical trials.
All the trials within a condition were identical to each other (basically, participants answered 20 times to the same item in the baseline condition and 50 times to another similar item in the experimental one).
I would run a generalized linear mixed model (since I have only 14 subjects... they are monkeys) but I have a different number of trials between the two conditions, and I'm not sure I can run a repeated measure analysis with such an experimental paradigm.
Any idea or suggestion?
thanks in advance for any help
all the best,
Marco
Essentially what you have here is an unbalanced design. Linear mixed-effects models have some robustness against unbalanced designs, and generalized linear mixed-effects models are possible. There are a number of useful LMEM packages for R, including lme4, whose glmer function can implement generalized models. The following paper gives a good overview of lme4 (though it's directed at psycholinguists): Article Fitting Linear Mixed-Effects Models Using lme4
Hi,
Try separately for each of the groups: General linear models repeated measures (SPSS). as the number of measures in follow-up is unequal. (Unbalanced).
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA