cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
JMP is taking Discovery online, April 16 and 18. Register today and join us for interactive sessions featuring popular presentation topics, networking, and discussions with the experts.
Choose Language Hide Translation Bar
adoucette
Level I

Please Help with Fisher's Exact Test?

Hi, I'm fairly new to JMP and have been using it for a small research project. (I'm just learning, and am a student not a statistician, but really like how intuitive the analyses are to run and the number and types of analyses supported.)

I'm struggling with putting together a Fisher's Exact Test for a small number of nominal data. I have 10 subjects who had their limb pulse measured before and after tourniquet application, then after two movements with the tourniquet on. I would like to compare pulse after application (all 10 are no pulse then) with pulse after movement 1 (7 are no pulse, 3 have pulse). But what the Y by X in JMP is giving me for Fisher's Exact test (and Chi Square) doesn't show any values for the RSquare, Pearson, or Fisher's Exact.

(online calculators like the one at socscistatistics.com do give me the data I'm looking for, but I can't seem to do it in JMP)

 

My data table looks like this:

Before TourniquetAfter Apply TourniquetAfter Move 1After Move 2
YesNoYesNo
YesNoNoYes
YesNoNoNo
YesNoNoNo
YesNoNoYes
YesNoNoYes
YesNoNoYes
YesNoYesNo
YesNoYesNo
YesNoNoYes

 

I've also tried setting it up like this:

 After ApplyAfter Move 1
Pulse Yes03
Pulse No10

7

 

But, again, getting the same thing in Y by X, which looks like the screenshot below.

How can / should I be setting this up so that I can get statistical data of pulse after move 1 vs pulse after apply?

Thanks for any assistance,

Ari

 

2020-10-17_12-37-55.png

 

Contingency(
	Y( :After Apply ),
	X( :After Move 1 ),
	Contingency Table,
	Fisher's Exact Test( 1 )
)
1 ACCEPTED SOLUTION

Accepted Solutions
statman
Super User

Re: Please Help with Fisher's Exact Test?

I'm not sure what question you are trying to answer?  Is the question: What is the affect of moves on pulse? Or does pulse change with moves? I attached the JMP data table as I think it might be setup.  I also added a script you can click on to see the analysis.

"All models are wrong, some are useful" G.E.P. Box

View solution in original post

3 REPLIES 3
txnelson
Super User

Re: Please Help with Fisher's Exact Test?

Your data has 2 dimensions, When the Tourniquet was applied, and Applied before or after a move.  Therefore, the data needs to be setup like the image below

Tourniquet After Move
Before Yes
After Yes
Before

No

Before Yes
After No
Before No
After Yes

The Fit Y by X platform can then be opened and the Tourniquet column used as the Y Response column, and the After Move used as the X Factor column

 

Jim
statman
Super User

Re: Please Help with Fisher's Exact Test?

I'm not sure what question you are trying to answer?  Is the question: What is the affect of moves on pulse? Or does pulse change with moves? I attached the JMP data table as I think it might be setup.  I also added a script you can click on to see the analysis.

"All models are wrong, some are useful" G.E.P. Box
adoucette
Level I

Re: Please Help with Fisher's Exact Test?

Thank you @txnelson and @statman. Both of the suggested solutions worked. You were correct that I needed to set the data table up a bit differently.

Statman, I'm testing if the pulse changes with moves. (So, for a tourniquet that initially made the pulse on that limb go away, return of pulse after the moves suggests that the moves may have caused the tourniquet to fail to make the pulse go away.)

Thanks again,

Ari