<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">I'm trying to understand the different statistical options in the study manger, when plotting channel measures.<div><br></div><div>I have data from 4 groups (3 different manipulations and one sham) that I gathered across sequential days of testing (e.g. Session 1,3,5), and my groups are unequal in size, e.g. 8,7,8,16 subjects (with 16 in sham group). Each 32-channel data set has 600-800 ERP trials in it, per subject and session.</div><div><br></div><div>In my Study design I have a Group x Session design set up, using (unpaired data) Group as the first variable (with all 4 values included) and (paired data) Session as the second variable, with just sessions 1 and 5 in the design.</div><div><br></div><div>Then I run the ERSP/ITC/ERP/Spectral precomputation, which takes a good long while (24 hours). </div><div><br></div><div>My questions:</div><div><br></div><div>1) Bootstrap and paired data: from what I gather I should not use bootstrap when looking at my Session data, so I've only looked at "Group" in this way. If I look at paired data (Session) could I just used an alpha of .025 and assume equivalence to a .05 level if the bootstrapping was working properly with paired data?</div><div><br></div><div>2) Non-bootstrap options: for permutation or parametric, and calculating stats on first and second variable, the matlab window output suggests that I should not be running this type of ANOVA on the parametric/permutation data, e.g.:</div><div><br></div><div>... (2 balanced 1-way ANOVAS and then 4 paired t-tests)</div><div>4 x 2, unpaired data, computing F values</div><div>Using balanced 2-way ANOVA<b><i> (not suitable for parametric testing, only bootstrap)</i></b></div><div>...</div><div><br></div><div>So is the plot channel measures allowing options to be selected that are not legitimate?</div><div><br></div><div>3) Use single trials (when available):</div><div>I understand from reading old list posts that the null hypothesis when using trial based stats is restricted to consideration of a specific subject's trials against the possible population of trials for that subject. What I don't understand is when I select "trials" with a multi-subject design, what is actually happening, e.g.: Did I just violate the assumptions of the analysis, or is this analysis now running at the subject level for my paired data, and correcting somehow to calculate/plot the stats for a study with many subjects? </div><div><br></div><div>For (3), when I use this in my study and calculate the Session variable differences in spectrum, many/most of the electrodes are significant when I check several 3-hz bins. When I don't use trial-based stats, almost none are (unless I un-check FDR). </div><div><br></div><div>4) FDR and Parametric: Unchecking FDR for Parametric seems plausible, since wouldn't that mcorrect be some sort of Bonferroni, which would be overly conservative given the highly correlated local information in ERP waveforms?</div><div><br></div><div><br></div><div>So.. can anyone help me clarify what kind of stats (using the Study manger) I should be setting up for this kind of study design? </div><div><br></div><div>Thanks,</div><div>Andrew</div><div><br></div></body></html>