Hi
This discussion - incl the subject line - really pains me. The truth of the matter is that your study at this stage does not give a conclusive answer to your research question. Rather than dodge this by fiddling with the analysis chain how about either getting more data (in case you;re lacking power given group sizes) or simply reporting it as is. As a reviewer I'd be much happier with either one of these options... even without reaching the nominal 5% corrected level if - as you say - your results speak to a convincing hypothesis clearly that is better than crow-baring things into shape...
hth
Christian
On 2 Nov 2012, at 08:59, Infinity <[log in to unmask]> wrote:
> Hi,
>
> I am doing a TBSS study with 30 subjects (15 of each group). After running randomize on the MD images I get corrected (1-p) values of 0.94 in regions that match my hypothesis. It's agonizingly close! Is there something I can do to boost these results?
|