Skip to main content
. 2022 May 23;12(10):1332. doi: 10.3390/ani12101332

Table 8.

Age as a predictor of testosterone. Individual, age range of analysis, age variable (age, age category), effect size with standard error (SE), Wald statistic, and p-value from GLMMs; and relative effect of age on mean testosterone concentrations. Degrees of freedom (df) is 1 in all pair-wise comparisons.

Individual Age Range of Analysis Age Variable Effect Size (SE) Wald p Age Effect
Male1 ~34–43 Age (years) −0.066 (0.021) 9.681 0.002 Decreasing
31–40 (reference) a
41–50 −0.097 (0.125) 0.606 0.436 -
Male2 32–53 Age (years) −0.027 (0.004) 37.043 <0.001 Decreasing
31–40 (reference) a
41–50 −0.289 (0.056) 26.262 <0.001 Lower
51–60 −0.432 (0.088) 23.858 <0.001 Lower
Male3 ~34–44 Age (years) −0.090 (0.015) 38.079 <0.001 Decreasing
31–40 (reference) a
41–50 −0.495 (0.082) 36.783 <0.001 Lower
Male4 b 15–32 Age (years) −0.017 (0.005) 10.509 0.001 Decreasing
11–20 (reference) a
21–30 −0.081 (0.063) 1.655 0.198 -
31–40 0.167 (0.128) 1.697 0.193 -

a Reference category for GLMMs against which other age categories are compared. b Age analysis includes only mature bulls, which started at age 15 for this individual.