Examples
Examples
Try the starting example
- Balanced crossed study JSON [part][operator][replicate] (measurement units)
- [[[9,11],[11,13]],[[19,21],[21,23]]]
- Repeatability variance (measurement units²)
- 2
- Operator variance, clamped (measurement units²)
- 2
- Part by operator interaction variance, clamped (measurement units²)
- 0
- Part variance, clamped (measurement units²)
- 50
- Combined gauge SD (measurement units)
- 2
Try another scenario
- Balanced crossed study JSON [part][operator][replicate] (measurement units)
- [[[9,11],[13,15]],[[19,21],[23,25]]]
- Repeatability variance (measurement units²)
- 2
- Operator variance, clamped (measurement units²)
- 8
- Part by operator interaction variance, clamped (measurement units²)
- 0
- Part variance, clamped (measurement units²)
- 50
- Combined gauge SD (measurement units)
- 3.162278
Calculation method
Calculation method
Repeatability variance (measurement units²)
MS_error from within part/operator cells.
Operator variance, clamped (measurement units²)
max(0,(MS_operator-MS_interaction)/(parts*repeats)).
Part by operator interaction variance, clamped (measurement units²)
max(0,(MS_interaction-MS_error)/repeats).
Part variance, clamped (measurement units²)
max(0,(MS_part-MS_interaction)/(operators*repeats)).
Combined gauge SD (measurement units)
sqrt(errorVariance+operatorVariance+interactionVariance).
Symbols and units
- a
- Balanced crossed study JSON [part][operator][replicate] (measurement units)
Use the units shown. A list keeps the order in which you enter its values. The symbol ^ means “raised to the power of”.
Assumptions & limits
Assumptions & limits
- Balanced random-effects two-way crossed ANOVA with interaction. Negative component estimates individually clamped to zero, without pooling or significance selection. No nested, destructive or unbalanced studies; not a certified gauge qualification. Supported limits: 2..20 parts; 2..10 operators; 2..20 identical replicate counts per cell; full rectangular balanced data.
Are these inputs private?
These calculations run in your browser. A shared link contains the values you choose to share. Only questions submitted to the separate AI finder are sent to OpenAI.
Further reading
Background reading; this does not imply endorsement or independent review of SumFizz.