Relative importance of predictor variables in explaining the variability in binomial migration (migrated/resided) general linear models and continuous straight line migration distance linear models.
| Response . | Predictor . | Type . | Levels . | N . | R2 . |
|---|---|---|---|---|---|
| Binomial | Release length | Cont. | 5 280 | 0.08 | |
| Time at liberty | Cont. | 5 273 | 0.06 | ||
| Size class | Disc. | 2 | 5 280 | 0.06 | |
| Region | Disc. | 9 | 5 332 | 0.06 | |
| Recapture length | Cont. | 2 919 | 0.05 | ||
| Decade | Disc. | 8 | 5 332 | 0.02 | |
| Year | Cont. | 5 332 | 0.01 | ||
| Sex | Disc. | 2 | 2 178 | <0.01 | |
| Continuous | Migrant/resident | Disc. | 2 | 5 332 | 0.61 |
| Time at liberty | Cont. | 5 273 | 0.11 | ||
| Release length | Cont. | 5 280 | 0.08 | ||
| Region | Disc. | 9 | 5 332 | 0.08 | |
| Size class | Disc. | 2 | 5 280 | 0.08 | |
| Recapture length | Cont. | 2 919 | 0.05 | ||
| Decade | Disc. | 8 | 5 332 | 0.04 | |
| Year | Cont. | 5 332 | 0.01 | ||
| Sex | Disc. | 2 | 2 178 | <0.01 |
| Response . | Predictor . | Type . | Levels . | N . | R2 . |
|---|---|---|---|---|---|
| Binomial | Release length | Cont. | 5 280 | 0.08 | |
| Time at liberty | Cont. | 5 273 | 0.06 | ||
| Size class | Disc. | 2 | 5 280 | 0.06 | |
| Region | Disc. | 9 | 5 332 | 0.06 | |
| Recapture length | Cont. | 2 919 | 0.05 | ||
| Decade | Disc. | 8 | 5 332 | 0.02 | |
| Year | Cont. | 5 332 | 0.01 | ||
| Sex | Disc. | 2 | 2 178 | <0.01 | |
| Continuous | Migrant/resident | Disc. | 2 | 5 332 | 0.61 |
| Time at liberty | Cont. | 5 273 | 0.11 | ||
| Release length | Cont. | 5 280 | 0.08 | ||
| Region | Disc. | 9 | 5 332 | 0.08 | |
| Size class | Disc. | 2 | 5 280 | 0.08 | |
| Recapture length | Cont. | 2 919 | 0.05 | ||
| Decade | Disc. | 8 | 5 332 | 0.04 | |
| Year | Cont. | 5 332 | 0.01 | ||
| Sex | Disc. | 2 | 2 178 | <0.01 |
Predictor variables used in each model: tagging and recapture length, size class at tagging (smaller or larger than 50 cm), tagging region, time at liberty, tagging decade (discrete), tagging year (continuous), and sex. Type of the response variable (continuous/discrete) is indicated in the Type column, number of levels for discrete variables in the Levels column, and number of observations in the N column. The R2 value is calculated as McFadden’s pseudo r-squared for the GLMs.
Relative importance of predictor variables in explaining the variability in binomial migration (migrated/resided) general linear models and continuous straight line migration distance linear models.
| Response . | Predictor . | Type . | Levels . | N . | R2 . |
|---|---|---|---|---|---|
| Binomial | Release length | Cont. | 5 280 | 0.08 | |
| Time at liberty | Cont. | 5 273 | 0.06 | ||
| Size class | Disc. | 2 | 5 280 | 0.06 | |
| Region | Disc. | 9 | 5 332 | 0.06 | |
| Recapture length | Cont. | 2 919 | 0.05 | ||
| Decade | Disc. | 8 | 5 332 | 0.02 | |
| Year | Cont. | 5 332 | 0.01 | ||
| Sex | Disc. | 2 | 2 178 | <0.01 | |
| Continuous | Migrant/resident | Disc. | 2 | 5 332 | 0.61 |
| Time at liberty | Cont. | 5 273 | 0.11 | ||
| Release length | Cont. | 5 280 | 0.08 | ||
| Region | Disc. | 9 | 5 332 | 0.08 | |
| Size class | Disc. | 2 | 5 280 | 0.08 | |
| Recapture length | Cont. | 2 919 | 0.05 | ||
| Decade | Disc. | 8 | 5 332 | 0.04 | |
| Year | Cont. | 5 332 | 0.01 | ||
| Sex | Disc. | 2 | 2 178 | <0.01 |
| Response . | Predictor . | Type . | Levels . | N . | R2 . |
|---|---|---|---|---|---|
| Binomial | Release length | Cont. | 5 280 | 0.08 | |
| Time at liberty | Cont. | 5 273 | 0.06 | ||
| Size class | Disc. | 2 | 5 280 | 0.06 | |
| Region | Disc. | 9 | 5 332 | 0.06 | |
| Recapture length | Cont. | 2 919 | 0.05 | ||
| Decade | Disc. | 8 | 5 332 | 0.02 | |
| Year | Cont. | 5 332 | 0.01 | ||
| Sex | Disc. | 2 | 2 178 | <0.01 | |
| Continuous | Migrant/resident | Disc. | 2 | 5 332 | 0.61 |
| Time at liberty | Cont. | 5 273 | 0.11 | ||
| Release length | Cont. | 5 280 | 0.08 | ||
| Region | Disc. | 9 | 5 332 | 0.08 | |
| Size class | Disc. | 2 | 5 280 | 0.08 | |
| Recapture length | Cont. | 2 919 | 0.05 | ||
| Decade | Disc. | 8 | 5 332 | 0.04 | |
| Year | Cont. | 5 332 | 0.01 | ||
| Sex | Disc. | 2 | 2 178 | <0.01 |
Predictor variables used in each model: tagging and recapture length, size class at tagging (smaller or larger than 50 cm), tagging region, time at liberty, tagging decade (discrete), tagging year (continuous), and sex. Type of the response variable (continuous/discrete) is indicated in the Type column, number of levels for discrete variables in the Levels column, and number of observations in the N column. The R2 value is calculated as McFadden’s pseudo r-squared for the GLMs.
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.
