Table IV.

Estimated G-study variance components for the (m:f)ri design and different D-study estimations

Subscale: G-study design:	Discipline (ni = 13)		(m:f)ri		Skill encouragement (ni = 4)
Source	df	σα2	(%)	σ(σα2)	Source	df	σα2	(%)	σ(σα2)
f	1	.0381	(5.6)	.0361	f	1	.000		.014
m:f	28	.049	(7.2)	.018	m:f	28	.138	(20.8)	.062
r	1	.030	(4.4)	.030	r	1	.001	(0.1)	.004
i	12	.172	(25.1)	.085	i	3	.014	(2.1)	.019
fr	1	.000		.003	fr	1	.000		.006
fi	12	.017	(2.5)	.013	fi	3	.000		.017
(m:f)r	28	.014	(2.0)	.009	(m:f)r	28	.005	(0.7)	.024
(m:f)i	336	.024	(3.6)	.016	(m:f)i	84	.202	(30.5)	.059
ri	2	.064	(9.4)	.030	ri	3	.000		.009
fri	12	.008	(1.2)	.010	fri	3	.000		.013
(m:f)rie	336	.267	(39.1)	.021	(m:f)rie	84	.304	(45.8)	.046
Total	779	.685				239	.664

D-study design:	Crossed2				Nested2				Crossed				Nested
nt	1	2	3	4	1	2	3	4	1	2	3	4	1	2	3	4
Mj	13	13	13	13	13	13	13	13	4	4	4	4	4	4	4	4
σ12	.049	.049	.049	.049	.049	.049	.049	.049	.138	.138	.138	.138	.138	.138	.138	.138
σδ2	.036	.019	.013	.011	.072	.037	.025	.019	.132	.091	.076	.071	.133	.092	.078	.071
Eρ2	.57	.72	.79	.82	.41	.57	.67	.72	.51	.60	.64	.66	.51	.56	.61	.64

Subscale: G-study design:	Positive involvement (ni = 32)		(m:f)ri		Problem-solving (ni = 32)
Source	df	σα2	(%)	σ(σα2)	Source	df	σα2	(%)	σ(σα2)
f	1	.002	(0.5)	.005	f	1	.001	(0.1)	.025
m:f	28	.052	(12.0)	.016	m:f	28	.389	(34.6)	.108
r	1	0		.001	r	1	.006	(0.5)	.007
i	31	.014	(3.2)	.007	i	31	.098	(8.7)	.032
fr	1	0		.001	fr	1	.000		.003
fi	31	.003	(0.8)	.005	fi	31	.001	(0.1)	.007
(m:f)r	28	.008	(1.8)	.004	(m:f)r	28	.036	(3.2)	.013
(m:f)i	868	.058	(13.4)	.012	(m:f)i	868	.135	(12.0)	.012
ri	31	.005	(1.0)	.005	ri	31	.034	(3.0)	.013
fri	31	.004	(0.8)	.006	fri	31	.001	(0.1)	.007
(m:f)rie	868	.286	(66.3)	.014	(m:f)rie	868	.424	(37.7)	.020
Total	1919				.432	1919	1.125

D-study design:	Crossed				Nested				Crossed				Nested
nt	1	2	3	4	1	2	3	4	1	2	3	4	1	2	3	4
ni	32	32	32	32	32	32	32	32	32	32	32	32	32	32	32	32
σr2	.052	.052	.052	.052	.052	.052	.052	.052	.389	.389	.389	.389	.389	.389	.389	.389
σδ2	.018	.010	.008	.006	.019	.010	.008	.006	.054	.029	.019	.017	.061	.045	.024	.015
Eρ2	.74	.84	.88	.90	.73	.83	.87	.89	.88	.94	.95	.96	.87	.93	.95	.96

Subscale:	Monitoring (ni = 11)
G-study design:	(m:f)ri
Source	df	σα2	(%)	σ(σα2)
f	1	.000		.131
m:f	28	1.618	(15.0)	.500
r	1	.853	(7.9)	.766
i	10	1.216	(11.2)	.774
fr	1	.000		.033
fi	10	.270	(2.5)	.215
(m:f)r	28	.004	(0)	.142
(m:f)i	280	.383	(3.6)	.364
ri	10	.726	(6.7)	.382
fri	10	.000		.159
(m:f)rie	280	5.722	(53.0)	.482
Total	1919	.432

D-study design:	Crossed				Nested
nr	1	2	3	4	1	2	3	4
ni	11	11	11	11	11	11	11	11
στ2	1.618	1.618	1.618	1.618	1.618	1.618	1.618	1.618
σδ2	.558	.298	.209	.166	1.478	.756	.516	.396
Eρ2	.74	.85	.89	.91	.52	.68	.76	.80

Note. f = fathers, m = mothers, r = raters, i = items. e = unmeasured facets that affect the measurement and/or random events. σ(σ²_α) = standard error of the variance component σ_α. The generalizability coefficients, E_ρ², were estimated based on pooled variance components across three sub-samples.¹ The variance component of the fixed father facet is likely biased. See endnote 1 for further elaboration. ² The crossed D-study design is formally designated as (m:f)RI, whereas the nested design is designated as (R:m:f)I.