Dear Lorenzo,
what about having three covariates: age, memory and their interaction
(after mean centering them, ie, (age-mean(age)).*(memory-mean(memory)) )
and contrasts [0 1 0 0 ], [0 0 1 0] and [0 0 0 1].
Best regards,
Guillaume.
On 01/07/16 13:40, lorenzo pasquini wrote:
> Hello
>
> I want to analyze the effect of age and of a memory scores on brain
> activity, and I am also interested on the interaction between age and
> memory on brain activity.
>
> I am not sure which statistical test I should use?
> 1. Option 1: a one-sample t test with two covariates (age and memory
> score). If I understand it correctly, the contrast would be [1 0] and [0
> 1] to test the effect of the single covariates. What is the contrast for
> the interaction?
>
> 2. Option2: A one-way anova, with a factor group and three levels for
> different ages (young, middle-aged, and old). Memory scores are added as
> a covariate. How should the contrast for the interaction between factor
> group and the covariate memory look like? Should I rather use a flexible
> factorial design?
>
> Thank you very much for your help and kind regards
>
> Lorenzo
>
>
>
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|