1
$\begingroup$

https://www.youtube.com/watch?v=gI0xQcLIplA

In the video link I have attached, a particular Friedman's test is implemented for a comparison of several treatments against a control. I am not sure if this implementation is well known. My knowledge is that Friedman compares several treatments in an all-to-all fashion. Also, I do not know if this tests maintains the Type I error constant.

This proposed test would be a non-parametric version of a Dunnett's test, with repeated measures.

Particularly, I wanted to know:

Regards, Ignacio

$\endgroup$
1
$\begingroup$

I have finally found theoretical discussion of the test in the following book (section 7.4).

http://www.amazon.com/Nonparametric-Statistical-Methods-Myles-Hollander/dp/0471190454

and this publication for extended tables. http://www.tandfonline.com/doi/abs/10.1080/03610917708812033

It is known as the multiple comparisons (Nemenyi, Wilcoxon-Wilcox, Miller) k - 1 treatments versus control test.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.