Table 3. Set of models used in a Cormack-Jolly-Seber recaptures-only analyses of cliff swallows to test hypotheses and estimate apparent survival and detection probability among years.^{* .}

Model	ΔQAICc	wi	K	-2LogLik	QDev
(1) φ(FY+year+g*Age), p(year+g*Age+RE) †	0.00	1.00	50	294863.8	937.81
(2) φ(FY+year+g*Age), p(year+g*Age)	116.04	0.00	49	295737.2	1055.85
(3) φ(FY+year+g*Age), p(year+net*Age)	150.34	0.00	47	296020.7	1094.16
(4) φ(FY+year+g*Age), p(year+RE)	166.96	0.00	45	296173.3	1114.77
(5) φ(FY+year+g*Age), p(year+unk*Age)	219.55	0.00	47	296532.9	1163.37
(6) φ(FY+year+g*Age), p(year+Age)	292.73	0.00	45	297104.0	1240.55
(7) φ(FY+year+g*Age), p(year)	301.25	0.00	44	297181.8	1251.07
(8) φ(FY+year+g*Age), p(year+sex)	303.32	0.00	45	297182.4	40159.78

^*Parameters with interactions are joined by '*', whereas parameters having parallel (additive) relationships are joined by '+'; Akaike's Information Criterion (AIC) values were corrected for over-dispersion (see Table 3), yielding quasi-AIC (QAIC_c) values; ΔQAIC_c values and model weights (w_i) were used to rank models; see Table 1 for model notations. Here “k” indicates the number of parameters in the model and “QDev” indicates the quasi-deviance of the model.

^†QAIC_c = 39946.48 for top-ranked model.