# Nonparametric repeated measures test to test for significant differences?

Some weeks ago, I used 5 different microphones to record the same sound. The idea was to compare what differences existed between these recordings. To do that I wanted to use some sort of test to check for significant differences in the audio features that I am interested (amplitude, fundamental frequency, etc.).

All the audio features from the recorded audios do not follow a normal distribution. For this reason, I should use nonparametric tests. Furthermore, since all the microphones recorded the same sound I should use some sort of repeated measures test. However, I am not sure what test I should use. I have only used repeated ANOVA before, but according to my knowledge that would not be correct for this case, as it is a parametric test.

Could you recommend me some non parametric test for repeated measures in order to control for differences? (it would be great that it was also implemented in R).

Friedman test maybe?

(microphones<-matrix(c(1:10,2:11,3:12,4:13,5:14), nrow=10)) # 5 microphones, 10 features
[,1] [,2] [,3] [,4] [,5]
[1,]    1    2    3    4    5
[2,]    2    3    4    5    6
[3,]    3    4    5    6    7
[4,]    4    5    6    7    8
[5,]    5    6    7    8    9
[6,]    6    7    8    9   10
[7,]    7    8    9   10   11
[8,]    8    9   10   11   12
[9,]    9   10   11   12   13
[10,]   10   11   12   13   14

friedman.test(microphones)

Friedman rank sum test

data:  microphones
Friedman chi-squared = 40, df = 4, p-value = 4.328e-08