topic Re: Chi-Square Analysis: WHERE is it in JMP? in Discussions

Chi-Square Analysis: WHERE is it in JMP?

Steven_Moore — Thu, 21 Aug 2014 16:26:00 GMT

Chi-Square Analysis: WHERE is it in JMP? Pretty simple in other software, but I would like to do it in JMP. I have data to determine if certain departments have disproportionately high injury rates compared to the overall rate. A goodness-of-fit exercise.

Re: Chi-Square Analysis: WHERE is it in JMP?

pmroz — Thu, 21 Aug 2014 17:42:46 GMT

Did you search the help in JMP? Found it in 2 seconds.

Help > Search the Help

Enter Chi-Square and hit Enter.

Re: Chi-Square Analysis: WHERE is it in JMP?

Steven_Moore — Thu, 21 Aug 2014 17:55:46 GMT

Yes, I did try the help, but did not get to what I was specifically looking for :(

Re: Chi-Square Analysis: WHERE is it in JMP?

pmroz — Thu, 21 Aug 2014 18:33:54 GMT

Can you post some of your data?

Re: Chi-Square Analysis: WHERE is it in JMP?

Steven_Moore — Thu, 21 Aug 2014 18:59:23 GMT

Here's what I am trying to get JMP to do for me (done in Excel): Calculated the Chi-Square Stats for each department and total:

Dept	Staff Level	Observed Injuries	Staff Proportion	Expected Injuries	Expected-Observed	(Expected-Obs)^2	Chi-Squared
A	53	1	0.125	7.63	6.63	43.9	5.76
B	12	5	0.028	1.73	-3.27	10.7	6.21
C	46	13	0.108	6.62	-6.38	40.7	6.15
D	78	14	0.184	11.22	-2.78	7.7	0.69
E	91	17	0.215	13.09	-3.91	15.3	1.17
F	37	3	0.087	5.32	2.32	5.4	1.01
G	29	3	0.068	4.17	1.17	1.4	0.33
H	29	3	0.068	4.17	1.17	1.4	0.33
I	49	2	0.116	7.05	5.05	25.5	3.62
Totals	424	61	1.000	61		Total Chi-Sq =	25.26

Re: Chi-Square Analysis: WHERE is it in JMP?

louv — Thu, 21 Aug 2014 20:11:51 GMT

smoore2,

Is this what you were trying to do?

Re: Chi-Square Analysis: WHERE is it in JMP?

Dan_Obermiller — Thu, 21 Aug 2014 20:54:11 GMT

I believe Lou has the proper approach. However, his example included the full raw data table for the different departments. Just in case you had summarized data (as presented in the sample data you provided), you would want to conduct your analysis using Frequency variables. I have included that example here.

Re: Chi-Square Analysis: WHERE is it in JMP?

Steven_Moore — Fri, 22 Aug 2014 11:05:18 GMT

Dan and Lou,

Thanks for your help. I was having difficult switching the method of data entry from the way I was doing with another software package.

Best regards,

Steven

Re: Chi-Square Analysis: WHERE is it in JMP?

louv — Fri, 22 Aug 2014 13:05:03 GMT

Steven,

No problem....Thank you for being so open to share the data. It was an excellent opportunity to refresh my understanding of JMP for your application.

Re: Chi-Square Analysis: WHERE is it in JMP?

David_Burnham — Fri, 22 Aug 2014 13:41:12 GMT

I actually think the format of the summary data in Steven's Excel spreadsheet is not that unusual. And I don't think it is always realistic to change the data entry format just to suit the idiosyncrasies of a particular analysis method. Which leaves (imho) contingency analysis languishing in Fit Y By X as the poor cousin to Bivariate and Oneway. It's tempting (especially for me) to write an add-in to deal with this problem, but it seems so basic and central to data analysis that JMP should deal with this. We have very powerful tools such as STACK and SPLIT that allow common data structures to be reshaped for particular methods of analysis. Is there not scope for a tool for reshaping of frequency-type data e.g. from Steve's format to DanO's ?

Dave

Re: Chi-Square Analysis: WHERE is it in JMP?

Dan_Obermiller — Fri, 22 Aug 2014 14:11:52 GMT

This is an interesting point. The key piece of information that is missing in Steven's data is the number of non-injured. So, if one were to add that column (a simple formula) and then stack the # injured column with the # of non-injured and you have the table ready for the analysis.

Re: Chi-Square Analysis: WHERE is it in JMP?

louv — Fri, 22 Aug 2014 14:28:16 GMT

Exactly why I chose to enter the data in the raw table format that I did include since from there it was easy to convert to Dan's table summary format using the Tabulate and Stack.

Re: Chi-Square Analysis: WHERE is it in JMP?

Steven_Moore — Fri, 22 Aug 2014 14:56:37 GMT

Dave@Pega_Analytics, thanks for your observation. You are right on, and that is why I was having a problem finding the Chi-Square Analysis I was looking for. In other software packages, {I dare not blashpheme here with the name of that software!!! ;) } this analysis is easier/quicker than in JMP.

Using what I call the "classical Chi-Square Analysis" that we learned in college stats class, there should be no need to include a column of non-injured. Each department (or classification) is assumed in the null hypothesis to produce a number of injuries at the same proportion as all the others. That's why I included the staffing levels for each department. Then the Chi-Square apporach calculates the statistic based on the expected and observed injuries only and a p-value is determined from the degrees of freedom and the alpha level.

Thanks for everyone's help.

Re: Chi-Square Analysis: WHERE is it in JMP?

Chris_Kirchberg — Fri, 22 Aug 2014 20:14:55 GMT

Not completely sure this is appropriate for the problem, but given that you are looking at a proportion and count data, using Fit Model>Generalized Linear Model might help. Set the Observed Injures and the Staff Level in the Y (as Continuous) then Dept as a Model effect. Choose Binomial Distribution and leave the Link Function as Logit.

The result is a series of ChiSquare tests, but not necessarily the same as the Contingency Analysis in Fit Y by X. Here we are testing if there a difference in the proportion of injuries to staff level per department. Whole model fit ChiSquare, Effect ChiSquare and Parameter Estimates (levels) ChiSquare's are calculated.

Goodness of Fit is not calculated. The Whole Model ChiSquared is the same as the Likelihood Ratio results in Fit Y by X. No Mosaic Plot either for communicating results graphically either.

Something to think about given the structure of your data.

Re: Chi-Square Analysis: WHERE is it in JMP?

BL — Tue, 25 Sep 2018 19:34:15 GMT

I get a network error when I try to download this and I am trying to do the chi square from summarized data.

Re: Chi-Square Analysis: WHERE is it in JMP?

txnelson — Tue, 25 Sep 2018 20:32:30 GMT

When you say "I get a network error when I try to download this", what is "this". One of the data tables attached to the discussions?