Hello everyone,
I am unsure about the method for reliability analysis of a scorecard.
I would like to analyze a scorecard in terms of its reliability. 5 different requests are evaluated using a scorecard by 4 different people each . Each of the 5 requests is given an overall score. I would like to investigate how "well" our scorecard performs in evaluating our requests.
One possibility would be to carry out the Measurement Systems Analysis in JMP. I could then calculate the percentage reproducibility using EMP Gauge R&R Results.
I have a few questions about this and hope someone can help me.
1) I am unsure whether this method is suitable for our experimental setup, since there are no repeat measurements.
2) In addition, our sample (4x5) is so small that the "Process Standard Variation" cannot be calculated. Can I still trust the results, and does it even make sense to carry out a Gauge R&R with such a small sample?
3) In addition to the "EMP Gauge R&R Results" output, there is also an "EMP Results" output. How do these outputs differ, and which one would be helpful for interpretation in my case?
4) How could I perform a Gauge R&R ANOVA in JMP?
5) Are there other better methods for analyzing the reliability of scorecards?
Thank you very much for your answers.
Best regards
Verena
