Use the Bias versus Reference Value plot to see how the bias values vary for each part. The blue dots represent the bias values for each reference value. The red square represents the average bias value for each reference value. The line is the least squares regression line fit to the average of the deviations.
Ideally, the deviations for each part are close to 0 and the fitted line is horizontal.
The coefficients are numbers from the regression line of the Bias versus Reference Value plot.
The general form of this least squares regression line is:
The term, b, represents the constant coefficient. It indicates where the fitted line crosses the y-axis.
The term, a, represents the slope coefficient. The slope of a line indicates the steepness of the line and is the change in the y-axis over the change in the x-axis.
When the slope coefficient, a, is very small, the slope is near horizontal. Thus, bias is relatively constant across reference values, and linearity is not a significant problem. Larger absolute values of the slope coefficient, |a|, indicate a steeper slope of the line. If the p-value of the slope is less than alpha, then linearity is significant.
In the absence of significant linearity, larger absolute values of the constant coefficient, |b|, indicate larger bias. When significant linearity is present, you must look at the individual bias values.
The standard error of the estimate of a regression coefficient measures how precisely the model estimates the coefficient's unknown value. The standard error of the coefficient is always positive.
Use the standard error of the coefficient to measure the precision of the estimate of the coefficient. The smaller the standard error, the more precise the estimate. Dividing the coefficient by its standard error calculates a t-value. If the p-value associated with the t-value is less than your α-level, you conclude that the coefficient is significantly different from 0.
S and R-Sq (R2) are measures of how well the model fits the data.
S is an estimate of σ, the standard deviation around the regression line.
R-Sq (R2) represents the proportion of variation in the bias that is explained by the linear relationship between the biases and the reference values.
Smaller values of S indicate less variability in the bias estimates. R2 ranges from 0 to 100%. Usually, the higher the R2 value, the better the model fits your data.
Linearity assesses the difference in average bias through the expected operating range of the measurement system. Linearity indicates whether your gage has the same accuracy (the same bias) across all reference values.
%Linearity is the linearity expressed as a percentage of the process variation.
To interpret the linearity of your data, determine whether the bias changes across the reference values. If the data do not form a horizontal line on a scatterplot, linearity is present. Ideally, the fitted line will be horizontal and will be close to 0.
For a gage that measures consistently across parts, the %linearity will be close to 0.
Bias is calculated as the difference between the known standard value of a reference part and the observed average measurement.The bias is a measure of a measurement system's accuracy.
%Bias is the bias expressed as a percentage of the process variability.
For a gage that measures accurately, the %bias is small.