Table 5 Model results showing the explanatory variables and their corresponding significance levels for all variables retained in the final models for periodic descents during the spawning and feeding period, e.g. the months August to September for the latter period. The response variable in both tests was the binomial variable periodic descents with a value of 1 representing fish that had performed this behaviour and 0 respesetning fish that had not performed the behaviour. The treatment contrast of R was used with the intercept value representing the value for females. Square brackets indicate the categorical variable sex, i.e. males, tested against this reference value. The top part of the model shows the results for the fixed effects, the bottom part shows the random effects.
Predictors | Periodic descents - sp | Periodic descents - fp | ||||
|---|---|---|---|---|---|---|
Log-Odds | CI | p | Log-Odds | CI | p | |
(Intercept) | -1.47 | -3.30–0.36 | 0.115 | -2.39 | -3.83 – -0.95 | 0.001 |
Length/sd(Length) | 0.19 | -0.13–0.51 | 0.240 | |||
Sex [Male] | -2.33 | -3.06 – -1.60 | < 0.001 | 0.39 | -0.62–1.40 | 0.450 |
Random Effects | ||||||
σ2 | 3.29 | 3.29 | ||||
τ00 | 0.32 Sp_Ground | 1.88 Sp_Ground | ||||
ICC | 0.09 | 0.36 | ||||
N | 7 Sp_Ground | 7 Sp_Ground | ||||
Observations | 320 | 135 | ||||
Marginal R2 / Conditional R2 | 0.290 / 0.352 | 0.007 / 0.368 | ||||