BookmarkSubscribeRSS Feed
Choose Language Hide Translation Bar
Grindel

New Contributor

Joined:

Apr 12, 2019

How to perform a Wilcoxon signed rank test for paired data

Hello everyone,

 

I'm very new to JMP and i'm exploring the possibilities with example found here and there on the net.

One thing i can't get to perform correctly is the Wilcoxon signed rank test.

 

I'm trying to follow the example given in this link  :
http://www.statstutor.ac.uk/resources/uploaded/wilcoxonsignedranktest.pdf

 

I'm going through "fit y by x" and selecting one column in y and one in x then non-parametric>Wilcoxon

 

The results is not the one given in the link nor the one calculated by hand.

It appaears the signed rank test is not the test that JMP is performing.

 

Capture.JPG

 

My suspicion is that it does not recognize paired data

 

Any help on this subjects would be greatly appreaciated

 

Have a great day,

 

 

1 ACCEPTED SOLUTION

Accepted Solutions
markbailey

Staff

Joined:

Jun 23, 2011

Solution

Re: How to perform a Wilcoxon signed rank test for paired data

It should look like this when you get it correctly:

 

Capture.PNG

Learn it once, use it forever!
3 REPLIES 3
markbailey

Staff

Joined:

Jun 23, 2011

Re: How to perform a Wilcoxon signed rank test for paired data

First, both variables must be numeric and continuous. (The Before variable is currently nominal.)

 

Second, select Analyze > Specialized Modeling > Matched Pairs. Be sure to enter the Before variable in the Y role, then the After variable.

 

See how that approach works.

Learn it once, use it forever!
markbailey

Staff

Joined:

Jun 23, 2011

Solution

Re: How to perform a Wilcoxon signed rank test for paired data

It should look like this when you get it correctly:

 

Capture.PNG

Learn it once, use it forever!
Highlighted
Grindel

New Contributor

Joined:

Apr 12, 2019

Re: How to perform a Wilcoxon signed rank test for paired data

Hello Mark,

 

Thank you for this quick and spot-on answer.

It seems so easy now that it's explained

 

With any luck it will help someone else in the future

 

Have a great day :)