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

Table 4 Model fit of confirmatory factor analysis in separate and combined groups (confirmatory sample)

Model	Model	N	Χ²	df	Χ²/df ratio	p	RMSEA	CFI	TLI
Autistic	0	155	35.92	5	7.18	.001	.2	.915	.829
Autistic	1	155	6.4	4	1.6	.171	.062	.993	.983
Possibly autistic	0	113	22.59	5	4.52	.001	.177	.924	.847
Possibly autistic	1	113	2.88	4	.721	.578	.001	1	1.01
Combined autistic/possibly autistic	0	268	52.46	5	10.49	.001	.189	.925	.85
Combined autistic/possibly autistic	1	268	4.36	4	1.09	.359	.018	.999	.999
Non-autistic	0	136	32.34	5	6.47	.001	.201	.927	.853
Non-autistic	1	136	18.32	4	4.58	.001	.163	962	.904
Non-autistic	2	136	2.87	3	.96	.412	.001	1	1.001

Recommended goodness of fit indices values demonstrating good model fit: χ²/df ratio close to zero (Bryant & Yarnold, 1995), RMSEA < 0.06, CFI > 0.95 and TLI > 0.9 (Browne, 2014; Hu & Bentler, 1993). Model 0 indicates baseline models without covaried error terms. Model 1 covaried error terms for items 1 + 5; Model 2 covaried error terms for items 3 + 4, and items 3 + 5 (Fig. 2)

ISSN: 2040-2392