more functionality?

[angelospsy]

more functionality?

[AngelosPsy]

Hey Christian,
Thank you for your nice comments!
Yes both great ideas! In principle we would have to go back to the literature and check how those things are analysed, and from there implement the choices here, but I can see that what can be done is just extend the models we have now to include more factors. The problem would be with the t-tests I think and how to combine the different factors (e.g., maybe take differences between factors and then compute t-tests) but what can be done is whenever we have more than 1 within subject factors to ignore the t-tests and run the other models. What do you think?
Best,
Angelos

[merzchris]

Hi, Angelos,

thanks for your quick reply! Going back to the literature would be one way, I agree, maybe more for the long run. But independent from this, an extension of the already existing models would also work, basically, the same analyses as currently implemented might be expanded to cover more than 2 CS or a factorial design. T-tests for CS+ vs. CS- could be expanded to repeated-measures ANOVAs including all CS and to use the t-tests as post hoc tests for the difference between all CS. The same would apply to the already existing ANOVAs. This might be more like an intermediate step, the first step you suggested (to ignore the t-tests and run the other models) would also work in the short run.

I hope these further ideas help a bit!

Best wishes,
Christian