Hi,
I have time-locked EEG data for 30 subjects who have gone through 2 experimental conditions (A and B). I also have a behavioral data from those subjects (X task scores). I wanted to know whether X scores differentially predicts EEG data for conditions A and B. Is it accurate to run ft_statfun_depsamplesregrT for this question? If so, does the resulting significant clusters indicate what I was looking for to learn or something else?
Thank you!
ANCOVA for two groups with covariate as behav. data.
Here is a detailed description in this link.
https://ncss-wpengine.netdna-ssl.com/wp-content/themes/ncss/pdf/Procedures/NCSS/Analysis_of_Covariance-ANCOVA-with_Two_Groups.pdf
Good afternoon Mustafa, I agree with Bhogaraju. It makes sense to run ANCOVA. But this will mostly only tell you how much variance is explained by the behaviour. If you specifically want to cluster you may consider a few different options. In SPSS you can do k-means clustering, setting a behavioural variable as the predictor. You could also try RapidMiner - it is an open source machine learning program that will let you load your data in .csv, specify the behavioural variable, and then it will conduct machine learning outcomes, and clustering, using your EEG data.
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix