Statistical analyses
All analyses were carried out using the statistical package R version 4.0.2 (R Development Core Team 2020). Data were tested for normality and homogeneity of variance using the Shapiro-Wilk normality test and residual plots respectively. Only the number of weavers foraging per diet per minute deviated from a normal distribution. To model relative diet preference, we built a generalized linear model with a Poisson error structure including diet, week, feeding session, sex and their two-way interactions as explanatory variables for variation in the number of weavers foraging on each diet per minute. We tested for differences in GUD between diets using a linear model. Diet, week, feeding session (time of day), and their two-way interaction terms were included in the model as explanatory variables for variation in GUD. To test the effect of diet restriction on body condition, we accounted for the effects of moult and aviaries on all indices by including moult status and aviary as predictor variables in the model for each body condition index. The explanatory power of both moult and aviaries on variation in all indices were not significant and so were dropped from all our models. Hence, we built linear models for each body condition index with diet, week, sex, and their two-way interactions as the main explanatory variables. For all models, a post hoc test was used to determine which groups differed significantly from each other.
RESULTS