where Yijk is the dependent variable, μ is the overall mean, Sourcei is the effect of DDGS sample, Levelj is the effect of level of DDGS inclusion, (Level × Source)ij is the interaction effect of DDGS sample and inclusion level, and eijk is the observational error for the (ijk)th observation. Because BW as a covariate was significant (P = 0.001), the fat pad was analyzed by an analysis of covariance using BW as the covariate.