Orthogonal polynomials
were used to determine linear and quadratic effects
of increasing feeding duration of CWG and soybean
oil. Contrast coefficients were derived using the integrative
matrix language (PROC IML) procedures of SAS.
The statistical model included block as the random effect
and sex, fat source, feeding duration, and all their
2-way and 3-way interactions as fixed effects. Hot carcass
weight was used as a covariate for last-rib backfat,
10th-rib backfat, loin eye area, and lean percentage.
Statistical significance and tendencies were set at P ≤
0.05 and P < 0.10, respectively, for all statistical tests.