topic Re: Gage R&R analysis? in Discussions

Gage R&R analysis?

statman — Sat, 10 Jun 2023 23:23:41 GMT

I need some help with my JMP friends.

I have attached a simple typical Gage R&R data set (3 operators, 5 samples and 3 repeated measures). This would be crossed then nested. When I run the analysis and get variance components, all seems OK. The I use the built in JMP Gage R&R platform (under the variability chart), the analysis is completely wrong. What am I doing wrong?

Re: Gage R&R analysis?

Mark_Bailey — Mon, 14 Dec 2020 21:35:53 GMT

Here is the result from the MSA platform:

Here is the result from the Variability Gauge platform with Gage R&R results:

Are these the results that you see, too?

Re: Gage R&R analysis?

Mark_Bailey — Mon, 14 Dec 2020 21:36:41 GMT

I think that the two factors are crossed. You do not enter Rep as a factor.

Re: Gage R&R analysis?

statman — Tue, 15 Dec 2020 18:17:34 GMT

@Mark_Bailey Mark, Here is what I see using the Gage R&R under Variability Plot:

Re: Gage R&R analysis?

Mark_Bailey — Tue, 15 Dec 2020 19:31:27 GMT

I agree that something is not correct. We are working with the same data table, but please confirm the meta-data shown on the left side of the data table:

I worry that column attributes or row states might have changed for you and are interfering with the analysis.

Re: Gage R&R analysis?

Mark_Bailey — Tue, 15 Dec 2020 19:36:59 GMT

Also, we should double-check the launch dialog. Here is the correct example:

Re: Gage R&R analysis?

Mark_Bailey — Tue, 15 Dec 2020 19:46:20 GMT

I also manually verified the calculations starting with the variance components shown above.

Re: Gage R&R analysis?

statman — Wed, 16 Dec 2020 00:49:07 GMT

@Mark_Bailey OK, I'll chalk that up to a SR. moment. I did not put sample in the spot that asks for the sample ID. You don't need to do that if you are just using the variance components. I mentioned I had never used the Gage R&R platform before (I'm not a fan of that approach to measurement system analysis though I realize it is used often, no offense intended).

Now on to more important topics. Mark & Pete, you guys are AWESOME. Thanks for your patience. Be safe!

Re: Gage R&R analysis?

Mark_Bailey — Wed, 16 Dec 2020 09:47:20 GMT

Please do not feel bad for this mistake. You provide invaluable help and guidance in the Community. The fact is that JMP is very complex and sophisticated. I have to look up things all the time. I was able to master JMP 2. Not so much with JMP 15!