R - Friedman test

[@chrislusk3497]

Very clear explanation thanks, and a very useful test considering how often the data don't fit the assumptions of repeated measures ANOVA.

[@stikpet]

Glad it was helpful!

[@NatashaKasher]

Thank you so much! Works like magic.

[@stikpet]

Glad it helped!

[@myrtopierrakou6145]

What to do if there are missing values?

[@stikpet]

if I'm not mistaken those are then simply ignored. The stats friedman.test function as an option na.action to change this, but am not familiar with that myself. Hope this helps.

[@sadatsalifu9701]

Very good video. Could you link to the csv file you used? Thanks.

[@stikpet]

yes, just now added the link in the description.

[@sadatsalifu9701]

@@stikpet Many thanks for this!

[@Danielmagox]

Thanks friend, very well explained, the effort is appreciated.

[@MarjoleinGroot]

thank you!

[@townofstrangers3831]

Can this test be used as a non-parametric alternative to two way repeated measures anova?

[@stikpet]

some say yes, some say no. You might want to look into a Scheirer-Ray-Hare test. Unfortunately I haven't looked into that myself yet.

[@townofstrangers3831]

@@stikpet Thank you!

[@preetikumar862]

Hi, this tutorial is so simple and easy to follow, however I keep getting an error saying Error in apply(y, 1L, rank) : dim(X) must have a positive length. i have tried all the alternatives given in the video. Any help would be great. Thank you.

[@stikpet]

did you download the script and follow it, or are you applying it to your own data? The dim(X) must have a positive length, might suggest that X is empty....

[@preetikumar862]

@@stikpet I was applying it to my own data. I rewatched the video and formatted my data according to yours and thank you it worked. This tutorial was extremely helpful with my college course work!

[@stikpet]

indeed the format of the data can sometimes be crucial. Glad it worked, and good luck with your analysis.

[@edouardgiudicelli2933]

Hello i dont understand your results. Your p is 5.1 isnt it superior to 0.5 ???

[@stikpet]

There is a e-09 at the end of that number, which is scientific notation for small numbers. Basically it means to divide that 5.1 by 10 to the power of 9, making it 0.0000000051 . Hope that clarifies.

[@edouardgiudicelli2933]

@@stikpet thank you very much

[@mjma1984]

@4:00, could not find function "friedman_test" !! I have all the packages installed ! is friedman_test or friedman.test ?

[@stikpet]

Did you also load the package with: library(coin)? The friedman_test is definately in the coin package, as listed in the documentation on page 82 of the reference manual at cran.r-project.org/web/packages/coin/coin.pdf.

[@zachlambe5110]

You are amazing thank you

[@stikpet]

glad it helped. On a side note, KZbin has cancelled my partner program since I don't have enough subscribers (currently around 715, should be 1000 😞), so it would really help if you subscribe (and all your friends) 😉

[@zachlambe5110]

716 now :)

[@lizadixon1741]

Saved a life.

[@stikpet]

glad it helped. On a side note, KZbin has cancelled my partner program since I don't have enough subscribers (currently around 642, should be 1000 😞), so it would really help if you subscribe (and all your friends) 😉

[@Super14alex14]

miss script

[@stikpet]

uhm, you mean you would like a link to the file I used in this video?

[@mateuspeixoto9620]

@@stikpet can be in the same R

[@stikpet]

Not really sure what you meant with your comment, but I've added a link to the script I used in the description of the video. Hope that this helps.

[@stikpet]

Not really sure what you meant with your comment, but I've added a link to the script I used in the description of the video. Hope that this helps.

[@stikpet]

Not really sure what you meant with your comment, but I've added a link to the script I used in the description of the video. Hope that this helps.