Confounding, Aliasing, and Resolution

Frac­tion­al DOEs enable us to effi­cient­ly test many fac­tors across vary­ing com­bi­na­tions in order to iden­ti­fy fac­tors wor­thy of deep­er inves­ti­ga­tion. How­ev­er, a trade­off for sim­plic­i­ty and speed is that we must be aware of the poten­tial for con­found­ing and alias­ing, or the inabil­i­ty to dis­tin­guish the inter­ac­tions between two or more fac­tors from each oth­er based on the exper­i­men­tal results. Learn how res­o­lu­tion lev­els play a role in address­ing con­found­ing, the five lev­els of res­o­lu­tion, and how to select the right res­o­lu­tion lev­el in our frac­tion­al exper­i­men­tal design to bal­ance the need for few­er runs with the abil­i­ty to inter­pret the results.