Linearity assesses the difference in average bias through the expected operating range of the measurement system.
Use the Bias versus Reference Value plot to see how the bias values vary for each part. For each reference value, the blue circles represent the bias values, and the red squares represent the average bias value.
The bias is the difference between a part's reference value and the operator's measurements of the part. The average bias values are also listed in the Gage Bias table under Bias.
A line is fitted through the bias values, using ordinary least squares regression. You want the data to form a horizontal line, which indicates that the bias does not change across the size of the parts, and that the measurement system does not contain significant bias. When the horizontal line is close to 0, the difference between the observed average measurement and the reference value is very small, which also indicates that the system does not contain significant bias
Generally, the closer the slope of the fitted line is to zero, the better the gage linearity. Ideally, the fitted line will be horizontal and will be close to 0.
The bias is the difference between a part's reference value and the operator's measurements of the part. Examine the %Bias for each reference value and the Average %Bias. The Average %Bias value is the bias expressed as a percent of the process variability.
For a gage that measures accurately, the %Bias value is also small.
Use the p-values used to test whether bias = 0 at each reference value and for the average bias.