We first compared absolute changes in daily time
budgets according to period (pre-rut and rut), fitted
as a categorical fixed effect using linear-mixed models
(LMM) with ‘male identity’ fitted as a random
term to control for repeated measurements of the
same individual (Pinheiro & Bates 2000). As year of
observation could have also influenced male activity
budgets, we included this variable as a categorical
fixed effect in our analyses. Because male age affects
time budgets of both species during and outside the
rut (Pelletier 2005; Mainguy & Coˆ te´ 2008), we also
included it as a continuous independent variable.
We used the overall proportion of time spent in each
of the five behavioural categories in each periodyear
in bighorn sheep as the sampling unit (n = 35
individual-year observations in both the pre-rut and
the rut), whereas in mountain goats we used yearly
individual means of the proportions of time computed
during activity budgets for each period
(n = 69 and 66 individual-year observations in the
pre-rut and the rut, respectively).