Specify the options for Evaluate Measurement Process (EMP Crossed)

Stat > Quality Tools > Gage Study > Evaluate Measurement Process (EMP Crossed) > Options
Alpha to calculate decision limits in ANOME and ANOMR plots
Enter the alpha value (also called significance level) for the ANOME and ANOMR plots. For the ANOME chart, the null hypothesis is that the average measurements among the operators are equal. For the ANOMR chart, the null hypothesis is that the average range among the operators are equal.
Usually, a significance level of 0.05 works well. A significance level of 0.05 indicates a 5% risk of concluding that the values are equal when they are not.
Use a higher alpha value, such as 0.10 or 0.25, if you want to be more likely to conclude that the operators measure differently. Use a lower alpha value, such as 0.01, to be less likely to conclude that the operators measure differently.
Process tolerance
Enter at least 1 specification limit to calculate probabilities that the variation in the measurement system causes misclassification of parts as good or bad.
  • Lower spec: Enter the lower specification limit.
  • Upper spec: Enter the upper specification limit.
Historical standard deviation
Enter a known value for the total variation, which is the part-to-part variation plus the measurement system variation. When you use the ANOVA method and enter at least 1 specification limit, then the calculations for the probabilities of misclassification use the historical standard deviation.
Measurement increment

The analysis compares the measurement increment to the probable error to determine whether the measurement increment is appropriate. By default, the analysis calculates the measurement increment from the data. When the analysis determines the increment and the measurements have decimals, the increment is 1 unit of the least, non-zero, place value in the measurements. For example, for the values 1.100, 1.400, and 1.900, the increment is 0.1. If the measurements do not have decimals, the analysis uses an increment of 1.

To report a valid conclusion in the results, specify the increment if the measurement increment from the data is inaccurate. For example, the gage that you use reports results in the tenths place, but records measurements only to the nearest half, such as 1.0 and 1.5. Then, specify that the increment is 0.5.