Dear everyone,
Please apologize for bringing up this topic again. Recently I had some discussions about conjunction analyses in the case of within-subject models like the Flexible factorial. I have to admit that I haven't really followed the last maybe one and a half years. Is there still no valid way to conduct conjunction analyses in instances like that due to the error term issue? In the particular case, it would be about conjunctions of A vs. B, A vs. C, A vs. D (experimental conditions vs. a control condition; contrasts could be specified on single-subject level or formed in the group model by incorporating the control condition as an additional level) and a conjunction of contrasts like B vs. C, B vs. D (contrasting experimental conditions).
Best regards
Helmut
|