# Statistic analysis on dependent and simultaneously independent data

Let's assume that I have a datasheet looking like that:

Patient Electrode  Task Theta power
1       1          1    0.5
1       2          1    0.8
1       3          1    0.1
1       4          1    0.9
...
1       1          2    0.3
1       2          2    0.9
1       3          2    0.4
1       4          2    0.3
...
2       1          1    0.2
2       2          1    0.7
2       3          1    0.3
2       4          1    0.4
...
2       1          2    0.6
2       2          2    0.7
2       3          2    0.8
2       4          2    0.8
...
3       1          1    0.0
3       2          1    0.3
3       3          1    0.2
3       4          1    0.8
...
3       1          2    0.2
3       2          2    0.5
3       3          2    0.1
3       4          2    0.9


I have for example 20 patients, each of the patient has 128 EEG electrodes recorded from one experiment with 2 categories. So data for the patients (for example this theta power on each electrode) are INDEPENDENT, because the data are from different people, and from only one run of the experiment. But on the other hand the data regarding electrodes from each patient are DEPENDENT, all electrodes share a bit of common signal.

What kind of statistical tests we can use to check the difference in for example this theta power between the task no. 1 and no.2? How to take care of this dependancy/independancy in any kind of data?

Of course, I'm dealing with it on my own way, but I wish to get to know your way of solving such problems.

• I will answer my own question then. I found intresting article on this topic. I encourage everyone who has the same problem to read that one ("Statistical testing in electrophysiological studies", Eric Maris, Psychophysiology, onlinelibrary.wiley.com/doi/epdf/10.1111/…)
– Mary
May 23, 2018 at 8:59
• I would suggest that your own answer is a little outdated. Check out a newer review of multiple methods (sciencedirect.com/science/article/pii/S0165027014002878) ... or even my own matlab-based toolbox for an actual implementation of the most successful method (github.com/Mensen/ept_TFCE-matlab)... these are really the only approaches when dealing with high-density EEG data as in your case. Jun 7, 2018 at 15:12
• @Mensen Thanks for that answer! I'm gonna check your toolbox then.
– Mary
Nov 19, 2018 at 12:28