The gage linearity table consists of the following:

·    Intercept and slope for the reference value versus bias plot.

·    Linearity, which measures how accurate your measurements are through their expected range. Ideally, linearity will be small in absolute value relative to the master part measurements.

·    %Linearity, which is the linearity expressed as a percent of the process variation. For a gage that measures consistently across parts, %linearity will be close to zero.

·    R, which is the proportion of the variation in the average bias that can be explained by its linear relationship with the master part measurements. If the R value is not close to 1, the relationship is not linear and linearity cannot be assessed. A non-linearity problem may exist.

For the parts data, the %linearity is relatively large (13.2%), which indicates a problem. Smaller parts tend to measure too high, while larger parts tend to measure too low.

Because the R value is high (71.4%), you can assume that the relationship between the master part measurement and bias is close to linear. It is reasonable to assess linearity for the parts data.