Minitab's regression bias method, which uses a t-test based on the fitted regression line to test the null hypothesis that bias is not present in the measurement system. A low bias value suggests that the attribute gage is measuring close to the reference values. Minitab also offers the AIAG method for testing bias if your situation meets the AIAG requirements of at least 20 trials per part and at least 6 parts greater than 0 number of acceptances. Because the regression and AIAG methods for testing bias use different calculations, the p-values may differ slightly for each method.

To determine if bias = 0, Minitab displays 3 statistics: a T-value (T), a p-value, and degrees of freedom (DF). The t and DF values are not very informative individually, but they are used to determine the p-value. The p-value tells you how likely it is that you would obtain the bias statistic if the null hypothesis (Ho) is true.

For the length data, bias is -0.0001045, T= -1.42655, DF = 6, and p = 0.203608. Because the p-value is greater than the chosen level of significance (a-level), typically set at 0.05, you can retain the null hypothesis that bias equals 0.