Hi @CLe_17,
Welcome in the Community !
The resolution of a design is a measure of the degree of confounding in the design. It's not an information displayed directly in the Evaluate Design platform, but you can check the Alias Matrix and Color Map on Correlations to check what is the resolution of your design.
Here is the resolution explanation from the JMP Help page :
- Resolution III - Main Effects Only : A resolution III design in which only main effects are estimable.
- Resolution IV - Some 2-factor interactions : A resolution IV design in which main effects and some two-factor interactions are estimable. The two-factor interactions are confounded.
- Resolution V - All 2-factor interactions : A resolution V design in which main effects and all two-factor interactions are estimable. Three-factor interactions are confounded with the two-factor interactions.
- Resolution V+ - All 2-factor interactions : A resolution V design in which main effects and all two-factor interactions are estimable. Some higher order interactions are estimable while others are confounded. The design is not a full factorial design.
- Resolution VI - Full Resolution : A full factorial design. There is no confounding of effects of any order.
Note that this definition is mainly for classical screening designs. For custom/optimal and modern designs, the aliasing structure may be more complex as the algorithm will try to avoid complete confounding between terms and will balance correlation between terms more "homogeneously", so the resolution may not make sense in this case (as no terms will be completely confounded).
It might be more interesting to look at correlation maps and alias matrix to evaluate how independantly can the terms be estimated, as well as other diagnostic metrics from Evaluate Design platform to better understand the design performances.
Hope this answer will help you,
Victor GUILLER
"It is not unusual for a well-designed experiment to analyze itself" (Box, Hunter and Hunter)
