A design point at which all the factors are run halfway between their high and low levels; all the factors must be continuous variables. Centerpoints can be used to add additional runs to experiments, to check for curvature in screening designs, and, if replicated, to estimate pure error. Also known as the design centerpoint.