nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

Huge DF denominators for repeated measures ANOVAs with JMP 14

I used to run repeated measures ANOVAs with JMP 13 with no proglems. But now I got JMP 14 and ran an ANOVA with the same data and somehow got "4326" as a DF Den under "Test Denominator Synthesis". This used to be "42" with JMP 13, and ovbiously the new F ratio got so much larger. I do not think there is anything different I did (data, how to set up), other than I am now using JMP 14 instead of JMP 13, and I really do not know what to do.

Could anyone please let me know what I am missing?

Thank you!

Accepted answers

All comments

Mark_Bailey

We do not know much about the details of your data or analysis but the increase in the DF for the denominator suggests that you have fewer DF for the model. Are there either fewer terms or fewer levels of your predictors?

Thank you very much for your response.

I attach a file with some images. Hopefully they will give you more information.

The image 1 shows how I specified the model with JMP 14, and the image 2 shows the output I get now. I am 100% sure that I used the same data set and 99% sure (I can no longer run JMP 13 to verify) that I used the same model specifiction, but I used to get an output like the image 3.

I do not know why DF Den are so different between the two versions, and would like to know how I could revert back to the output I used to get with JMP 13. I would really appreciate your help. Thank you!

RepeatedMeasures.docx

Mark_Bailey

The fact that your results with JMP 13 exhibit fractional degrees of freedom for the denominator of the F-test suggests that you used REML to estimate the parameters and variances. REML is the default (and recommended) procedure so that makes sense. On the other hand, EMS is selected for the analysis done with JMP 14. Try using REML with JMP 14.

Thank you very much for your response.

I tried using REML, but the results are the same as the attached image.

I am wondering if you could try running my data. It would really help me if you could. I will attach my data set with the following message. Thank you!

REML JMP14.PNG

Here is the data file attached.

>Thank you very much for your response.

>I tried using REML, but the results are the same as the attached image.

>I am wondering if you could try running my data. It would really help me if you could. I will attach >my data set with the following message. Thank you!

MUdata.xlsx

Mark_Bailey

I verified your results with JMP 14.

It appears to me that almost half of the terms, all higher-order interactions, may be removed from the model.

I am not surprised that there are 4326 degrees of freedom for the denominator (MSE) of the F-tests. You start with 4576 observations and your model contains 208 parameters. I do not understand how the DF could have been only 42 in the original analysis.

Mark_Bailey

It would have been helpful if you had suppled the JMP data table. I had to restore the meta-data and the model in Fit Least Squares by hand.

Mark_Bailey

I don't see the need to nest Subject within Disease. I verified that the results without nesting are identical.

Thank you again for the quick reply!

I am sorry about the Excel vs. JMP formats, and I can see that for REML (although it makes a difference for EMS), there is no need for nesting Subject within Disease.

But my original questions still stay:

1. Why are the outputs so different between JMP 13 and JMP 14? If you look at the Test Denominator Synthesis sections for both, the JMP 13 secion lists different DF Den's for different factors, while the JMP 14 section list just one number (4326) for all the factors except Disease (42). Also, the Denom MS Synthesis column lists different factor combinations for different factors for JMP 13 but just "Residual" for all the factors except Disease for JMP 14.

2. How could I revert back to the output I used to get with JMP 13? I would like to use the JMP 13 version, as it matches the results from SPSS and Unix-ranova.

Thank you again.

Mark_Bailey

They are not different. Not at all.

I copied the data from the Excel workbook to a JMP data table. I set up the same full factorial linear model with Subject as the random effect in Fit Model dialog. I used REML to fit the data with JMP 14. I saved the script for the Fit Least Squares analysis and then saved the data table. I closed the data table and JMP 14. I launched JMP 13, opened the data table and ran the script. The results are identical.

I then submitted the Fit Model job to SAS, which uses PROC MIXED. The results are identical. (Except for the different parametrization in the SAS procedure.)

So the DF for the denominators in question are the same (4326) by all three analyses (JMP 13, JMP 14, and SAS 9.4).

Also, as I mentioned before, these DF seem reasonable given the number of observations and the number of parameter estimates.

Thank you again for your quick response.

I now have to appologize. I am sorry I made a mistake in remembering the model I used to use. I found that my old laptop still had JMP 13 (which will be good until this March), started playing with it, and finally remembered that I used to put in all possible combinations of the factors including Subject as the attached image shows. In this way, I can get the output I showed you as Image 3 of my RepeatedMeasures.docx file. If I try to use REML instaed of EMS, JMP 13 complains that there is not enough memory. I also tried to run the same model with JMP 14 on my new laptop, but then I get the "Unable to allocate enough memory" error message for both EMS and REML.

I like this output as it gives the same results as Unix-ranova and SPSS (General Linea Model>Repeated Measures>Full factorial, SS Type III). I could try to do all the analyses with JMP 13 on my old laptop for these couple of months as a quick fix, but I do still have altered versions of my original questions as below:

1') Why putting in all possible factors change DF Den's so much (different DF Den's for different factors vs. just one number for all the factors except Disease, and for the Denom MS Synthesis column, different factor combinations for different factors vs. just "Residual" for all the factors except Disease)?

2') Ideally I would like to continue running my ANOVAs with JMP 14 on my new laptop, but would there be any way to allocate more memory?

Thank you.

JMP13Model.JPG