The Inferential Statistics Menu from ELLISTAT It includes five entries, but not all of them are equally suitable for the same purpose. Three of them fall under classical statistics: they allow you to test a hypothesis formulated in advance, such as «Does this new setting reduce the variance?».
The last two work the other way around: they do not impose any initial assumptions; instead, they help build one by comparing the problematic pieces with those that are not problematic.
This distinction changes the way you approach the menu: before choosing an appetizer, you first need to decide whether you’re looking to confirm a hunch or to form one.
| Entrance | What is it used for? |
| Test | Test a hypothesis using your grid data. This is the main entry point. |
| Test on Summary Statistics | Perform a test without the raw data, using the sample size, the mean, and the standard deviation. |
| Power | Designing an experiment: How many measurements are needed to detect a given deviation? |
| BOB vs WOW | Compare the best and worst products to see what sets them apart. |
| Component Inversion | Identify, by swapping components, which one is causing the variation. |
💡 The first three are traditional statistical tools. The last two are methods of root cause analysis : They do not test a pre-established hypothesis; rather, they help formulate it.
