You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I want to conduct a mediation analysis for EEG data, and the X variable would be conditions, Y would be reaction times, and Mediator would be amplitudes in each electrodes. But here comes the problem: since there are more than 50 electrodes (i.e. more than 50 mediators) and only 1 X and Y variable, how can I conduct this mediation analysis?
Here's my tentative solution, but I am not sure whether it is possible: Can I conduct mediation analysis for each electrodes, and select those electrodes that have significant mediation effect?
I would really appreciate it if anyone could help me.
The text was updated successfully, but these errors were encountered:
Hi everyone,
I want to conduct a mediation analysis for EEG data, and the X variable would be conditions, Y would be reaction times, and Mediator would be amplitudes in each electrodes. But here comes the problem: since there are more than 50 electrodes (i.e. more than 50 mediators) and only 1 X and Y variable, how can I conduct this mediation analysis?
Here's my tentative solution, but I am not sure whether it is possible: Can I conduct mediation analysis for each electrodes, and select those electrodes that have significant mediation effect?
I would really appreciate it if anyone could help me.
The text was updated successfully, but these errors were encountered: