Enter your data for Attribute Gage Study (Analytic Method)

Stat > Quality Tools > Gage Study > Attribute Gage Study (Analytic Method)

Enter your data

Select the option that best describes your data.

Summarized counts

If you have observed counts that are summarized by category, complete the following steps.

  1. In Part numbers, enter the column that contains the part names or numbers.
  2. In Reference values, enter the column that contains the reference values for each part. The reference value is the known standard measurement for each part.
  3. Under Binary attributes in, select Summarized counts. Organize summarized data in the worksheet so that each row contains the part name or number, reference value, and summarized count. Parts can be text or numeric. Reference values and counts must be numeric.
  4. In Number of trials, enter the number of times that each part was measured.
  5. In Attribute label, you can enter a custom label.
In this worksheet, Part contains the part number identifiers, Reference contains the known standard measurement, and Acceptances contains the summarized count of acceptances of that part.
C1 C2 C3
Part Reference Acceptances
1 1.35 0
2 1.40 3
3 1.45 8
4 1.50 13
5 1.55 15
6 1.60 18
7 1.65 19
8 1.70 20

Raw data

If you have a column of binary categorical (raw) data, complete the following steps.

  1. In Part numbers, enter the column that contains the part names or numbers.
  2. In Reference values, enter the column that contains the reference values for each part. The reference value is the known standard measurement for each part.
  3. Under Binary attributes in, select Raw data. Organize raw data in the worksheet so that each row contains the part name or number, reference value, and binary response. Parts and responses can be text or numeric. Reference values must be numeric.
  4. In Attribute for calculation, enter the attribute for estimating probabilities, which must match one of the two binary response values. For example, if you have the data values of pass and fail, enter either "pass" or "fail" depending on your desired calculation reference.
In this worksheet, Part contains the part number identifiers, Reference contains the known standard measurement, and Response contains the accept or reject status of each run.
C1 C2 C3
Part Reference Response
1 1.35 Reject
1 1.35 Reject
1 1.35 Reject
1 1.35 Reject
8 1.70 Accept
8 1.70 Accept
8 1.70 Accept
8 1.70 Accept

Tolerance limit for calculation

Enter either the lower limit or the upper limit that you want to use for the bias and repeatability calculation.
  • Lower limit: Enter the lower tolerance limit (also called lower specification limit).
  • Upper limit: Enter the upper tolerance limit (also called upper specification limit).
Display the other limit (Optional)
Although you need to specify one only tolerance limit to run the analysis, you can enter the other tolerance limit to display it on the Gage Performance Curve. When an attribute gage has a lower and upper limit, the gage performance curves are mirror images if you can assume linearity and uniformity of error. If the gage measures differently at each end of the tolerance, such as with a go/no-go pin gage that has two separate pins, perform a separate analysis for both upper and lower conditions.