2

I want to find out whether a neural measure measured at 12 time-points is correlated with a behavioral measure measured on the same time-points. I have data from 20 subjects.

I'm hoping to perform a permutation test where I compare the value of the Pearson correlation coefficient between the 12 neural and behavioral data points in the actual order, with a null distribution generated by permuting the 12 time-points and re-calculating the Pearson correlation 10,000 times.

Since the analysis above is performed on each subject's data, I would like to know how to combine the results across the 20 independent subjects.

mdewey
  • 16,541
  • 22
  • 30
  • 57
Marcelo Mattar
  • 97
  • 7

1 Answers1

2

Combining independent $p$-values is a method which has been used in meta-analysis. Two common methods (Fisher's method and Stouffer's method) are outlined in https://en.wikipedia.org/wiki/Fisher%27s_method There are other methods which are discussed in the chapter by Becker in

Cooper, H and Hedges, L V
A handbook of research synthesis 1994 (Russell Sage, New York)
mdewey
  • 16,541
  • 22
  • 30
  • 57