Table 1.

Model fit indices for the different BAT models.

Model		χ2	S-Bχ2	df	CFI	TLI	RMSEA [90% CI]		Δχ2	p
Core symptoms
1	Unidimensional model	2960.13	1.29	230	0.73	0.70	0.13 [0.12–0.13]
2	Correlated 4-factor model	773.54	1.28	224	0.95	0.94	0.06 [0.05–0.06]	2 vs. 1	1700.46	<0.0001
3	Second-order model (4 first order, 1 second order)	776.79	1.28	226	0.95	0.94	0.06 [0.05–0.06]	3 vs. 1 3 vs. 2	1522.52 3.25	<0.0001 0.20
Secondary symptoms
4	Unidimensional model	1963.62	1.29	104	0.70	0.65	0.15 [0.15–0.16]
5	Correlated 2-factor model	852.11	1.30	103	0.88	0.86	0.10 [0.09–0.10]	5 vs. 4	5482.03	<0.0001
5a	Adjusted correlated 2-factor model	500.48	1.29	102	0.94	0.92	0.07 [0.07–0.08]	5a vs. 4 5a vs. 5	2413.39351.63	<0.0001
Core & secondary symptoms
6	Correlated 6-factor model	2244.29	1.25	687	0.91	0.90	0.06 [0.05–0.06]
7	Second-order model (6 first order, 1 second-order)	2309.48	1.25	696	0.91	0.90	0.06 [0.05–0.06]	7 vs. 6	65.19	<0.0001
8	Second-order model (6 first order, 2 second-order)	2293.77	1.25	695	0.91	0.90	0.06 [0.05–0.06]	8 vs. 6 8 vs. 7	49.4815.71	<0.0001 <0.0001

Note. χ² = chi-square; S-Bχ² = Satorra–Bentler scaling factor for chi-square; df = degrees of freedom; CFI = comparative fit index; TLI = Tucker–Lewis index; RMSEA = root mean square error of approximation; Δχ² = difference in chi-square; Δdf = difference in degrees of freedom; p = p-value. The factor-loading matrices of the models and all correlations between the latent variables are available upon request from the first author.