<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 12pt;
font-family:Calibri
}
--></style></head>
<body class='hmmessage'><div dir='ltr'><font style="font-size:12pt;" face="Times New Roman" size="3">Thanks for the quick reply!<br>However, when I tried it I got an "Reference to non-existent field 'erpdata'" error.<br>Not really sure why, though.<br><br>Deniz<br id="FontBreak"></font><br><br><br><div>> Date: Fri, 29 Aug 2014 09:48:48 -0700<br>> Subject: Re: [Eeglablist] p and F values, difference wave, mean amplitudes<br>> From: mdschall@ucsd.edu<br>> To: deniz.dohmen@live.de<br>> CC: eeglablist@sccn.ucsd.edu<br>> <br>> Hi Deniz,<br>> <br>> If you are computing T-tests then you don't need an F value, but you'll<br>> find your T value the same way through EEGLAB. The only way I know how is<br>> through the command line - you can reference this part of the wiki for<br>> more information<br>> http://sccn.ucsd.edu/wiki/Chapter_08:_Command_line_STUDY_functions (under<br>> plotting stats & receiving results subsection)<br>> <br>> If you are more comfortable using the graphic interface, set up all your<br>> stat testing parameters first in the GUI, then rather than clicking the<br>> plot ERPs button, run this command:<br>> [stats, df, pvals, surrog]=statcond(STUDY.changrp(n).erpdata) - where n=<br>> the numerical index of the channel of interest. You can find the T-values<br>> in the variable called stats. Stats has the same number of data points as<br>> your ERP variable, which are samples, not time points. So if you want to<br>> see the T-value at latency 300 msec, check out the variable<br>> STUDY.changrp(n).erptimes. Times are stored in seconds, so look for 0.300,<br>> and note the column number. Go back and type stats(column number), and<br>> you'll have the T-value at that time. Do the same for pvals(column number)<br>> to find the corresponding p value at that time point.<br>> <br>> Hope this helps!<br>> <br>> best,<br>> Matt<br>> <br>> PhD Candidate, Cognitive Science, UCSD<br>> Adjunct Professor, Psychology, Chapman University<br>> https://sites.google.com/site/hazmattneurolab/<br>> <br>> <br>> > Dear List,<br>> ><br>> > I did all the preprocessing of my data, epoched and created a STUDY with 4<br>> > conditions for 11 subjects (I'm only comparing conditions a vs b and c vs<br>> > d, respectively. Thus, two paired t-tests).<br>> > I also get the graphs and a grey shaded area for the significant<br>> > latencies. However, I need to get t and p values for my masterthesis.<br>> > Really struggling to get those.<br>> > Furthermore, I need to get the mean amplitude of the respective difference<br>> > waves for the 350-450ms latency (in order to correlate them with another<br>> > measure.<br>> > Any advice on how to go about this?<br>> > I'm under a bit of time pressure and would appreciate any help!<br>> ><br>> > Cheers,<br>> > Deniz<br>> ><br>> > _______________________________________________<br>> > Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html<br>> > To unsubscribe, send an empty email to<br>> > eeglablist-unsubscribe@sccn.ucsd.edu<br>> > For digest mode, send an email with the subject "set digest mime" to<br>> > eeglablist-request@sccn.ucsd.edu<br>> <br></div> </div></body>
</html>