Development and validation of the suicidal behaviours questionnaire - autism spectrum conditions in a community sample of autistic, possibly autistic and non-autistic adults

Table 6 Results of tests for invariance in SBQ-ASC across gender and visual aids in the combined autistic/possibly autistic group, and non-autistic group

Model	Χ²	df	Model fit	CFI	TLI	ΔM	Model difference		P
Model	Χ²	df	RMSEA	CFI	TLI	ΔM	Δdf	Δχ²	P
Comparison of gender (combined autistic/possibly autistic group)
M1: Configural invariance (unconstrained)	8.62	8	.015	.999	.998				.375
M2: Metric invariance	9.7	11	.001	1	1	M2–M1	3	1.08	.782
M3: Scalar invariance	11.03	12	.001	1	1	M3–M2	1	1.32	.25
M4: Residual invariance	12.02	16	.001	1	1.01	M4–M3	4	.999	.91
Comparison of visual aids (combined autistic/possibly autistic group)
M1: Configural invariance (unconstrained)	5.58	8	.001	1	1.01				.694
M2: Weak factorial/metric invariance	7.31	11	.001	1	1.01	M2–M1	3	1.73	.631
M3: Scalar invariance	7.45	12	.001	1	1.01	M3–M2	1	.147	.7
M4: Strict invariance	13.73	16	.001	1	1	M4–M3	4	6.27	.18
Comparison of gender (non-autistic group)
M1: Configural invariance (unconstrained)	19.38	6	.093	.98	.933				.004
M2: Weak factorial/metric invariance	23.68	9	.08	.978	.951	M2–M1	3	4.302	.231
¹M3: Scalar invariance	24.28	10	.074	.979	.957	M3–M2	1	.595	.44
M4: Strict invariance	42.46	14	.089	.957	.939	M4–M3	4	18.18	.001
Comparison of visual aids (non-autistic group)
M1: Configural invariance (unconstrained)	15.01	6	.075	.987	.955
M2: Weak factorial/metric invariance	18.31	9	.062	.986	.969	M2–M1	3	3.297	.348
M3: Scalar invariance	18.41	10	.056	.987	.975	M3–M2	1	.105	.746
M4: Strict invariance	20.54	14	.042	.99	.986	M4–M3	4	2.132	.711

RMSEA = Root-mean-square error of approximation. CFI = comparative fit index. TLI = Tucker-Lewis index
¹Marginally significant degradation in fit is seen after this model (increase in RMSEA > .015 and reduction in CFI > .01)

ISSN: 2040-2392