[Eeglablist] p and F values, difference wave, mean amplitudes

Matthew Schalles mdschall at ucsd.edu
Fri Aug 29 09:48:48 PDT 2014


Hi Deniz,

If you are computing T-tests then you don't need an F value, but you'll
find your T value the same way through EEGLAB. The only way I know how is
through the command line - you can reference this part of the wiki for
more information
http://sccn.ucsd.edu/wiki/Chapter_08:_Command_line_STUDY_functions (under
plotting stats & receiving results subsection)

If you are more comfortable using the graphic interface, set up all your
stat testing parameters first in the GUI, then rather than clicking the
plot ERPs button, run this command:
[stats, df, pvals, surrog]=statcond(STUDY.changrp(n).erpdata) - where n=
the numerical index of the channel of interest. You can find the T-values
in the variable called stats. Stats has the same number of data points as
your ERP variable, which are samples, not time points. So if you want to
see the T-value at latency 300 msec, check out the variable
STUDY.changrp(n).erptimes. Times are stored in seconds, so look for 0.300,
and note the column number. Go back and type stats(column number), and
you'll have the T-value at that time. Do the same for pvals(column number)
to find the corresponding p value at that time point.

Hope this helps!

best,
Matt

PhD Candidate, Cognitive Science, UCSD
Adjunct Professor, Psychology, Chapman University
https://sites.google.com/site/hazmattneurolab/


> Dear List,
>
> I did all the preprocessing of my data, epoched and created a STUDY with 4
> conditions for 11 subjects (I'm only comparing conditions a vs b and c vs
> d, respectively. Thus, two paired t-tests).
> I also get the graphs and a grey shaded area for the significant
> latencies. However, I need to get t and p values for my masterthesis.
> Really struggling to get those.
> Furthermore, I need to get the mean amplitude of the respective difference
> waves for the 350-450ms latency (in order to correlate them with another
> measure.
> Any advice on how to go about this?
> I'm under a bit of time pressure and would appreciate any help!
>
> Cheers,
> Deniz
>
>  		 	   		  _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to
> eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu




More information about the eeglablist mailing list