TABLE 3.
Incident ratios (95% confidence interval (CIs)), percentage of independent effects of variable contribution to the model (%IE), and overall percentage of deviance explained from top‐ranked generalized linear models (GLMs) calculated for Pink Salmon from Hogan Bay (Hogan) and Stockdale Creek (Stockdale) even‐year lineage sampled in 2014
| Stream | Sex | Length (mm) | %IE | Origin | %IE | Date (day) | %IE | Distance (m) | %IE | Intertidal | %IE | % Deviance |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Hogan | Females | 1.01 (0.99–1.02) | 3% | 0.42 (0.24–0.71) | 65% | 1.03 (0.98–1.07) | 6% | NA | NA | 0.59 (0.33–1.05) | 26% | 6% |
| Hogan | Males | 1.01 (1.00–1.02) | 60% | NA | NA | NA | NA | 0.997 (0.994–1.000) | 40% | NA | NA | 4% |
| Stockdale | Females | 1.00 (1.00–1.01) | 1% | 0.60 (0.45–0.79) | 20% | 0.97 (0.94–1.00) | 5% | 0.998 (0.998–0.998) | 74% | NA | NA | 25% |
| Stockdale | Males | 1.02 (1.01–1.02) | 16% | 0.43 (0.31–0.60) | 28% | 0.95 (0.91–0.98) | 12% | 0.998 (0.997–0.998) | 44% | NA | NA | 36% |
Explanatory variables include: “Length (mm)” = parent mideye to fork length, “Origin” = parent origin (hatchery‐ vs. natural‐origin), “Date (day)” = parent sample date as day of year, “Distance (m)” = parent sample location in terms of distance above the upper extent of the intertidal, and “Intertidal” = categorical variable of parent sample location within or above the upper extent of the intertidal. Incident ratios were derived from the models with the best fit, as determined by Akaike Information Criteria (AIC). Note that incident ratios are presented in the units of each variable. Variables not included in a model are indicated by NA.