If you do not have observations for enough factor level combinations, you cannot create a multi-vari chart.
To create a multi-vari chart, you must include observations for at least 60% of the possible combinations of factor levels. Each possible combination of factor levels is a cell.
Suppose you are studying two factors, A and B. Factor A has three levels and Factor B has two levels. There are six possible combinations of factor levels (cells) in this design. If data are not included for one of these combinations, it creates an empty cell (*).
C1 | C2 | C3 |
---|---|---|
A | B | Observations |
A | 1 | * |
B | 1 | 20 |
C | 1 | * |
A | 2 | 22 |
B | 2 | * |
C | 2 | 21 |
A | 1 | * |
B | 1 | 23 |
C | 1 | * |
... | ... | ... |