How do I interpret Gauge R&R Results?

[#1]

I have two questions about Gauge R&R results:

1. If I calculate a Gauge R&R value with 6*StdDev, do I have to report the measurement uncertainty as +-3*StdDev with confidence interval of 99.73?
2. Since JMP fits some models to calculate the different components, is there any error with these values? The reason I ask is because I have to two different sets of data, and I would like to know if the difference I see is statistically significant.

[#2]

Correct, the precision of your measurement system is +/- 3*StdDev if you want to have it contain 99.73% of your measurement error distribution.  You can adjust the reported X*StdDev depending on the criticality of the measurement error.

If you right click on the Variance Components for GRR box you can ask for the 95% confidence levels.

[#3]

Thanks for your answer, it is very useful.

One more question: the variance components table has only Reproducibility, whereas the Gauge R&R table splits the Reproducibility in two parts, operator and operator*part.

Is it possible to get the variance component for these two parts together with 95% Lower and 95% Upper?