[Eeglablist] p-value and t-test on ERPs

[Makoto Miyakoshi]

Dear Haneen,

There are a fixed effect model and a random effect model.
https://urldefense.com/v3/__https://www.meta-analysis.com/downloads/Meta-analysis*20Fixed-effect*20vs*20Random-effects*20models.pdf__;JSUlJQ!!Mih3wA!TY9TJrs6DBZ9cp6l5NTDE-IFijLxY07676mMaxbZxpUJkhuVWl6BseRKdXLlD1dCWjuWXw$ 


That being said, traditionally, in your case you take individual mean first
and run the t-test across subjects.

Makoto

On Wed, Jul 1, 2020 at 1:50 AM Haneen Alsuradi <haneen at nyu.edu> wrote:

> Hi EEGLAB list,
>
> I have a question on calculating the p-value of ERPs under two conditions,
> c1 and c2 (for a specific channel). I have 20 subjects, each performed 50
> trials for c1 and performed 50 trials for c2. I have 500 data point for
> each trial. My aim is to find the time points/time period where there is a
> significant difference between the neural activity under the two
> conditions. So my plan was to conduct 500 two sample t-test between the two
> conditions.
>
> However, I am not sure, should I take the average of the 50 trials per
> subject and then conduct the t-test or should I give the 50 trials as is
> without averaging and conduct the t-test?
>
> Will appreciate your feedback and insight.
>
> Thanks
> Haneen
> _______________________________________________
> 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
>