Table 1.

Akaike information criterion (AIC) from two candidate models applied to Gulf of Alaska Pacific ocean perch larvae (n = 1084). The log-likelihood (logL), AICc and ΔAICc measures determine the Akaike weights (w) which indicate the model with the greatest relative support.

candidate model	fixed effects	random effect	logL	AICc	ΔAICc	w
linear mixed effects	age, week, year, week*year	female	4177	−8329	0	1
linear regression	age, week, year, week*year		3836	−7660	669	0