Table 5.

Final GLMMs (binomial distribution) of factors explaining variance in P₂ in the two female populations. Pure crosses are set as intercepts and therefore P‐values refer to the difference between the respective factor and the intercept. P values for cross × time interactions have been Bonferroni corrected (i.e., multiplied by 3) to account for multiple comparisons (comparison of cross × time interaction of each cross to that of the control cross). Bolded P‐values indicate that the difference in time trend between the pure cross and the respective cross is statistically significant. Crosses are abbreviated by female population, 1st male, and 2nd male population, respectively: C, Colorado; V, Vancouver

Effect	Parameter estimate	SE	z‐value	P
Intercept (CCC cross)	1.01	0.33	3.1	0.002
CCV cross	−0.36	0.54	−0.7	0.50
CVC cross	−0.75	0.48	−1.6	0.12
CVV cross	−0.70	0.53	−1.3	0.19
time	−0.31	0.11	−2.8	0.005
CCV × time	0.30	0.18	1.6	0.30
CVC × time	0.57	0.16	3.6	0.001
CVV × time	0.47	0.18	2.6	0.03
Intercept (VVV cross)	1.30	0.36	3.6	<0.001
VCC cross	0.19	0.48	0.4	0.68
VCV cross	1.25	0.46	2.7	0.007
VVC cross	−1.20	0.49	−2.4	0.014
time	−0.23	0.12	−1.9	0.06
VCC × time	−0.02	0.16	−0.1	1
VCV × time	−0.66	0.16	−4.2	<0.001
VVC × time	0.81	0.17	4.7	<0.001