“High level” means without getting into the details.
“Group” is related to the measurement error. With a large measurement error any single part will have a large area of measurements values due to the lack of repeatability. So if we measure 2 parts, part A and part B, A might be less than B. If we measure them again A may be larger than B. This range of measurement error defines how large the ‘group’ is. It might be helpful to think of this as a range or a cloud of error. This range exists around each measurement - it is not a fixed interval on the value axis.
While it is admirable that you want to understand this, I think we may have a language issue here and without diving into the math it is best to understand that this calculation adds no value to understanding the measurement system.
If you really want more detailed information about this I suggest reading Donald Wheeler’s An Honest Gauge R&R.
Note: I disagree with Dr. Wheeler regarding the usefulness of the Intraclass Plot (aka ISoPlot or more accurately the Youden Plot). I have found it to be much more useful than any mathematical assessment of measurement error. I have a paper on this in the resources section.
“Group” is related to the measurement error. With a large measurement error any single part will have a large area of measurements values due to the lack of repeatability. So if we measure 2 parts, part A and part B, A might be less than B. If we measure them again A may be larger than B. This range of measurement error defines how large the ‘group’ is. It might be helpful to think of this as a range or a cloud of error. This range exists around each measurement - it is not a fixed interval on the value axis.
While it is admirable that you want to understand this, I think we may have a language issue here and without diving into the math it is best to understand that this calculation adds no value to understanding the measurement system.
If you really want more detailed information about this I suggest reading Donald Wheeler’s An Honest Gauge R&R.
Note: I disagree with Dr. Wheeler regarding the usefulness of the Intraclass Plot (aka ISoPlot or more accurately the Youden Plot). I have found it to be much more useful than any mathematical assessment of measurement error. I have a paper on this in the resources section.
