Hi Steve,
Re: the first question. Yes, that's what I was asking, and that confirms what I've been doing so far, thanks!
Re: the second question. I'm still not clear on exactly how one would run the correlation I am looking for. My experience has been using FEAT for correlations, and entering demeaned values as a separate EV. Given that I only want to use the t-values that fall in my Randomise clusters (i.e. bypassing FEAT, cluster thresholding), what would be the most efficient way to look at correlations?
Thanks,
Anna
|