Skip to main content
. 2023 Sep 11;13:14967. doi: 10.1038/s41598-023-42166-2

Table 2.

Results of final generalised linear mixed models, with Poisson error variance and log-link function and binomial error variance and logit-link function, examining effects of age (as a linear and quadratic term), age at first reproduction (AFR), age at last reproduction (ALR), environmental conditions (NDVI), and an interaction of age (as a linear and quadratic term) and NDVI on the number of recruits and recruitment probability per female per year.

Source of variation Response variables
Recruit number (N = 2422 broods of 1802 females) Recruitment probability (N = 2422 broods of 1802 females)
Intercept − 1.366 (− 1.602, − 1.129)*** − 1.244 (− 1.613, − 0.875)***
Age 0.112 (− 0.012, 0.236)’ 0.187 (− 0.027, 0.400)’
Age2 − 0.060 (− 0.109, − 0.010)* − 0.095 (− 0.181, − 0.009)*
AFR − 0.034 (− 0.106, 0.038) − 0.069 (− 0.193, 0.055)
ALR 0.187 (0.097, 0.278)*** 0.225 (0.070, 0.381)**
NDVI − 0.071 (− 0.223, 0.081) − 0.117 (− 0.356, 0.122)
Residualsfledgling number model 0.975 (0.877, 1.074)*** 1.326 (1.180, 1.471)***
Fixed termspooled 1.001 1.844
Female IDrandom 0.000 0.000
Plot IDrandom 0.077 0.249
Yearrandom 0.192 0.422
Residual 1.323 3.290

Residuals from the fledgling number model were entered as a covariate to control for the number of produced offspring. To reduce models, only non-significant interactions were removed if P > 0.10. For fixed terms, parameter estimates accompanied by 95% CIs are presented. Variance estimates for fixed terms (pooled), each random factor, and the residual component obtained from the final model are also shown. The significance level for fixed terms is coded as ***(P < 0.001), **(P < 0.01), *(P < 0.05), and ’(P < 0.10). Significant terms are in bold.