How to use Friedman's test as one-way ANOVA instead of two-way?

Hi,

I have 15 subjects and I measure a parameter in them over time (4 different times). So I have a matrix A (15x4) and I would like to test the significance of change over time. It means that I have 4 groups with 15 observations in each of them. If I wanted to go with parametric methods, I would have chosen one-way repeated measures ANOVA. As my parameter does not have a normal distribution, I have to use a non-parametric method to detect a significant change over time. So I use friedman(A,1) assuming that I only have one replicate of the parameter for each subject and the results that I get are reasonable. Can someone please confirm if this is the correct way of using friedman?

Thanks

Answer 1

[Aditya]

Hi Pegah,

Yes, using the Friedman test in this context is appropriate. The Friedman test is a non-parametric alternative to the one-way repeated measures ANOVA and is suitable for analyzing data from a repeated measures design when the assumption of normality is violated.

Here is the documentation link for the same:

https://www.mathworks.com/help/stats/friedman.html