. Author manuscript; available in PMC: 2009 Nov 1.

Published in final edited form as: J Educ Psychol. 2008 Nov;100(4):829–850. doi: 10.1037/a0012657

Table 4.

Fit indices and Model Comparisons for the Competing Measurement Models

Model	df	χ²	p	NFI	NNFI	CFI	SRMR	Δχ² Model 1
Pretreatment Measurement Models
Six-Factor Model
1. L, A, DA, C, NVR, WP	64	58.31	.6771	.965	1.000	1.000	.0389
Five-Factor Models
2. A, C, NVR, WP, L+DA	69	145.88	.0000	.927	.995	.965	.0783	87.49^*
3. L, C, NVR, WP, A+DA	69	199.78	.0000	.897	.912	.933	.1325	91.47^*
4. L, A, NVR, WP, C+DA	69	147.54	.0000	.926	.952	.964	.0739	89.23^*
5. L, A, C, WP, NVR+DA	69	155.82	.0000	.917	.946	.954	.1803	97.51^*
6. L, A, C, NVR, WP+DA	69	137.19	.0000	.931	.960	.970	.0686	78.88^*
One-Factor Model
7. General Factor	77	634.07	.0000	.745	.731	.775	.1753	575.76^*
Posttest Measurement Models
Two-Factor Models
1. Near Transfer, Far Transfer	4	6.67	.1541	.981	.982	.993	.0323
2. Simple, Complex	4	92.61	.0000	.729	.472	.736	.1456	84.94^*
One-Factor Model
3. General Math Factor	5	83.41	.0000	.686	.384	.692	.1423	76.74^*

Note. N = 122. For pretreatment models, L = language factor; A = attention factor; DA = dynamic assessment factor; C = calculation factor; NVR = nonverbal reasoning factor; S = story problem factor. NFI = norm fit index; NNFI = nonnormed fit index; CFI = comparative fit index; SRMR = standardized root-mean-square residual.

p < .001.