The effects of group size and subtherapeutic antibiotic alternatives on growth performance and morbidity of nursery pigs: a model for feed additive evaluation

Olsen, Kristin M; Gabler, Nicholas K; Rademacher, Chris J; Schwartz, Kent J; Schweer, Wesley P; Gourley, Gene G; Patience, John F

The UNIVARIATE procedure of SAS (SAS Inst. Inc. Cary, NC) was used to determine homogeneity of variances and to identify outliers. Observations were considered outliers if greater than three SD from the mean. Residual plots were also used to verify equality of variances and normality of the residuals. It was determined that all variables analyzed met the assumptions for parametric tests, so the same model was used to analyze all the data. The MIXED procedure of SAS was used to analyze the data with pen as the experimental unit and initial BW as a covariate. The fixed effects were diet, group size, and diet × group size interaction. Block was considered a random effect. Differences were considered significant if P < 0.05 and tendencies if 0.05 ≥ P < 0.10.