Definition
Measurement system analysis (MSA) determines how much of the observed variability is caused by the measurement itself. A repeatability and reproducibility study (Gauge R&R) splits variability into parts, instrument (repeatability) and operators (reproducibility). A measurement system is acceptable when its share of total variability or tolerance stays around 10 %; up to 30 % is conditionally usable.
Quality control & SPC
Without a verified measurement neither control charts nor capability indices make sense — they may describe gauge noise instead of the process. A typical study: 10 parts × 3 operators × 2–3 repeats, evaluated by the average-and-range method or ANOVA, which can reveal the operator × part interaction.
For attribute inspection (pass/fail) an appraiser agreement study (kappa) is used. Gauge linearity and stability over time are assessed as well. Results are expressed as %R&R relative to tolerance and the number of distinct categories (ndc ≥ 5).
In Statistica
Statistics → Industrial Statistics & Six Sigma → Process Analysis contains Gauge R&R by the range and ANOVA methods, linearity and stability studies, attribute MSA (kappa) and plots of variance components, operator × part interactions and R charts by operator; it generates the study design together with a data-collection sheet.
Related terms
- Process capability (Cp, Cpk)Capability indices compare process variability with the customer's tolerance limits.
- ANOVA (analysis of variance)ANOVA (analysis of variance) tests whether the means of three or more groups differ.
- Standard deviation and varianceVariance is the average squared deviation of values from the mean; the standard deviation (SD) is its square root and has the same units as the data.
- Six SigmaSix Sigma is a data- and statistics-driven methodology for process improvement.
Knowledgebase guides
FAQ
- What %R&R is acceptable?
- Up to 10 % acceptable, 10–30 % conditionally depending on cost and importance of the characteristic, above 30 % unacceptable — improve the gauge before further decisions.
- Range method or ANOVA?
- ANOVA is more precise and reveals the operator × part interaction; the average-and-range method is simpler and still common in the automotive industry (AIAG).
Try it on your own data
Statistica free for 30 days
Full version, no credit card. Or get a pricing estimate in a minute.
Quality control & SPC
Updated: September 2026.