thanks christian for your comments.
yes you understood perfectly. actually my template was calculated from a larger population of controls (46), and then I tested patients vs just 30 controls.
you are telling me that if I want to use a controls only template (as I suspect that my destroyed patients RSN could ruin my group analysis) it must be done using a controls population much larger (but how much ???) than those used to group comparison or "small" but totally different from the compared one ?
thanks
Alberto
|