Linear Mixed Effects Models
Linear mixed effects models including bird identification number and cohort number as random factors were used to analyze the longitudinal plasma data to test the hypotheses (H1a) that acute flight decreases oxidative capacity during flight and enables oxidative damage levels to remain low, and (H1b) that regularly flying increases antioxidant capacity and oxidative damage. Linear mixed models that controlled for experimental cohort as a random factor were also used to test the hypothesis that flight-training (H1c), dietary fat (H2), and dietary antioxidants (H3) influenced Δ during training of plasma lipid damage and lipid damage in the liver and pectoralis. We conducted pairwise comparisons using least square means to determine the differences in oxidative status between specific timepoints.