The hemorrhagic transformation index score: a prediction tool in middle cerebral artery ischemic stroke

BMC Neurology

Table 8 CFA. Goodness of Fit Statistics

	F1-V4	F1-V5	F1-V6	F1-V7	F2	Description
Likelihood ratio
χ²(2/5/9/14/13)	0.5	56.17	87.26	103.66	18.19	Model vs. saturated
p-value	0.78	<0.001	<0.001	<0.001	0.15
χ²(6/10/15/21/21)	594.14	664.18	711.01	727.4	727.4	Baseline vs. saturated
p-value	<0.001	<0.001	<0.001	<0.001	<0.001
Satorra-Bentler test
χ²(2/5/9/14/13)	0.4	50.96	82.63	100.31	18.08	Model vs. saturated
p-value	0.82	<0.001	<0.001	<0.001	0.16
χ²(6/10/15/21/21)	475.75	574.14	649.06	684.79	684.79	Baseline vs. saturated
p-value	<0.001	<0.001	<0.001	<0.001	<0.001
Population error
RMSEA	<0.001	0.138	0.127	0.109	0.027	Root mean squared error of approximation
90% CI	0–0.056	0.107–0.172	0.104–0.153	0.09–0.13	0–0.054
p-close	0.932	<0.001	<0.001	<0.001	0.909	Probability RMSEA ≤0.05
RMSEA-SB	<0.001	0.131	0.124	0.107	0.027	Satorra-Bentler RMSEA
Information criteria
AIC	3726.82	4682.79	5446.18	6121.68	6038.21	Akaike information criterion
BIC	3778.21	4747.02	5523.26	6211.6	6132.42	Bayesian information criterion
Baseline comparison
CFI	1	0.92	0.89	0.87	0.99	Comparative fit index
TLI	1	0.84	0.81	0.81	0.99	Tucker-Lewis index
CFI-SB	1	0.92	0.88	0.87	0.99	Satorra-Bentler CFI
TLI-SB	1	0.84	0.81	0.81	0.99	Satorra-Bentler TLI
Size of residuals
SRMR	0.01	0.08	0.08	0.08	0.03	Standardized root mean squared residual
CD	0.83	0.83	0.83	0.83	0.96	Coefficient of determination

Note: F1-V4 indicates one-factor model with 4 variables (ASPECTS, NIHSS, HMCA sign, and AF on ECG), F1-V5 one-factor model with 5 variables (4 previous variables + HR on ECG), F1-V6 one-factor model with 6 variables (5 previous variables + sex), F1-V7 one-factor model with 7 variables (6 previous variables + INR), and F2 two-factor model with factor 1 (ASPECTS, NIHSS, and HMCA sign) and factor 2 (AF on ECG, INR, sex, and HR on ECG)

ISSN: 1471-2377