The gage bias table consists of:

·    Bias for each reference value and average bias, which is the average of the differences between the actual measurements and the reference part measurement for each part.

·    %Bias for each reference value and average %Bias, which is the bias expressed as a percent of the process variability.

·    P-values used to test whether bias = 0 at each reference value and for the average bias.

Ideally, the bias will be close to zero relative to the reference part measurements.

·    A positive bias indicates that the gage overestimates.

·    A negative bias indicates that the gage underestimates.

For a gage that measures accurately, the %bias will be small as well.

For the parts data, the average bias (-0.053333) and %bias (0.4) indicate that there is some variability due to bias. The p-value is 0.040. Because the p-value is less than the chosen level of significance (a-level), typically set at 0.05, you can reject the null hypothesis that bias equals 0.