Table 3.

Predictive models of offspring sex ratio with a variable composition of the male segment in 10 experimental treatments in Kaamanen, Finland. (Models are logistic regression on the general form y=a+b(x). s.e., standard error; G, a test statistics for model fit that is ∼χ² distributed; P model, test for overdispersion in the model; AICc, Akaike information criterion corrected for small sample size; ΔAICc, difference in AICc value relative to the best model (bolded); AICc−w, AICc weights. The AICc−w in line for the different hypotheses (in italics) are the summed values for parameters related to the specific hypothesis.)

parameter (x)	a	s.e.(a)	b	s.e.(b)	Z	P	residual deviance	d.f.	G	P (model)	AICc	ΔAICc	AICc−w
differential allocation hypothesis													0.894
mean male age	−0.6333	0.2703	0.2495	0.1036	2.409	0.016	5.118	8	0.640	1.000	51.855	0.999	0.236
mean male mass	−1.2167	0.4680	0.0126	0.0048	2.606	0.009	4.120	8	0.515	1.000	50.856	0.000	0.389
mean sire age	−0.5118	0.2587	0.1871	0.0921	2.032	0.042	6.854	8	0.857	0.999	53.590	2.734	0.099
mean sire mass	−0.9972	0.4391	0.0094	0.0041	2.276	0.023	5.779	8	0.722	0.999	52.516	1.660	0.170
Trivers–Willard model													0.047
mean female age	−0.9791	1.3713	0.1952	0.2789	0.700	0.484	10.523	8	1.315	0.995	57.260	6.404	0.016
mean female mass—September	1.0988	2.0726	−0.0141	0.0261	−0.541	0.588	10.720	8	1.340	0.995	57.457	6.601	0.014
mean female mass—October	2.0699	2.5901	−0.0267	0.0330	−0.808	0.419	10.359	8	1.295	0.996	57.096	6.240	0.017
various other factors													0.059
number of males	0.0680	0.1700	−0.0140	0.0222	−0.631	0.528	10.614	8	1.327	0.995	57.351	6.495	0.015
adult sex ratio (proportion of males)	0.0375	0.1623	−0.6548	1.4716	−0.445	0.656	10.815	8	1.352	0.995	57.552	6.696	0.014
mean birth date	−2.0091	3.6150	0.0143	0.0260	0.550	0.582	10.711	8	1.339	0.995	57.448	6.592	0.014
proportion of females calving	1.0128	1.4483	−0.0113	0.0157	−0.716	0.474	10.501	8	1.313	0.995	57.237	6.381	0.016