Skip to main content
. Author manuscript; available in PMC: 2012 Feb 1.
Published in final edited form as: J Abnorm Psychol. 2011 Feb;120(1):198–209. doi: 10.1037/a0021660

Table 3.

Confirmatory Factor Analysis of 25 Axis I and Axis II Disorders: Fit of Models

Model χ2 df p (Δχ2 test) CFI TLI RMSEA
1. Only method factors 1,364.46 77 .69 .65 .109
2. One factor 568.85 96 <.001 .89 .90 .059
3. Two factors 424.50 95 <.001 .92 .93 .050
4. Three factors 393.55 93 <.001 .93 .93 .048
5. Four factors 313.14 93 <.001 .95 .95 .041
6. Four factors, three cross-loadings 247.64 91 <.001 .96 .96 .035

Note. For Sample 2, N = 1,399. All models included two orthogonal method factors, and the observed variables were regressed on sex and age simultaneously with testing of the factor model. With the weighted least squares mean and variance adjusted (WLSMV) estimator, the chi-square and degrees of freedom were estimated rather than calculated. The appropriate DIFFTEST procedure in Mplus was applied to perform the Δχ2 test. CFI = comparative fit index; TLI = Tucker–Lewis index; RMSEA = root-mean-square error of approximation.