Table 5.
Fit Indices for Latent Growth Curve Models
| Model | AIC | BIC | Chi square | RMSEA [CI] | CFI | SRMR |
|---|---|---|---|---|---|---|
| Social Support | ||||||
| Unconditional intercept-only | 1050.75 | 1068.23 | χ2 (4) = 6.38, p = .17 | .049 [.000, .118] | .994 | .098 |
| Unconditional linear | 1050.37 | 1078.35 | χ2 (1) = .008, p = .93 | .000 [.000, .054] | 1.000 | .001 |
| Model 1 | 1051.75 | 1093.71 | χ2 (3) = 1.75, p = .63 | .000 [.000, .088] | 1.000 | .011 |
| Model 2 | 1049.61 | 1098.57 | χ2 (4) = 2.40, p = .66 | .000 [.000, .076] | 1.000 | .011 |
| PTSD Symptoms | ||||||
| Unconditional intercept-only | 1280.98 | 1298.46 | χ2 (4) = 10.93, p = .03 | .084 [.026, .146] | .990 | .057 |
| Unconditional linear | 1279.30 | 1307.28 | χ2 (1) = 3.26, p = .07 | .096 [.000, .221] | .997 | .012 |
| Model 1 | 1234.48 | 1276.44 | χ2 (3) = 4.66, p = .20 | .048 [.000, .127] | .998 | .011 |
| Model 2 | 1228.52 | 1277.48 | χ2 (4) = 5.83, p = .21 | .043 [.000, .113] | .998 | .009 |
| Satisfaction with Life | ||||||
| Unconditional intercept-only | 2215.07 | 2232.56 | χ2 (4) = 1.50, p = .83 | .000 [.000, .057] | 1.000 | .014 |
| Unconditional linear | 2219.79 | 2247.77 | χ2 (1) = .22, p = .64 | .000 [.000, .132] | 1.000 | .006 |
| Model 1 | 2218.16 | 2242.64 | χ2 (8) = 6.66, p = .57 | .000 [.000, .066] | 1.000 | .019 |
| Model 2 | 2214.56 | 2242.53 | χ2 (10) = 7.04, p = .72 | .000 [.000, .052] | 1.000 | .017 |
| Reintegration Difficulty | ||||||
| Unconditional intercept-only | 1363.22 | 1380.70 | χ2 (4) = 7.00, p = .13 | .055 [.000, .122] | .995 | .055 |
| Unconditional linear | 1362.88 | 1390.86 | χ2 (1) = .66, p = .41 | .000 [.000, .157] | 1.000 | .007 |
| Model 1 | 1357.89 | 1382.37 | χ2 (8) = 7.41, p = .49 | .000 [.000, .071] | 1.000 | .037 |
| Model 2 | 1347.91 | 1375.89 | χ2 (10) = 9.07, p = .53 | .000 [.000, .065] | 1.000 | .033 |
Note. PTSD = post-traumatic stress disorder. AIC = Akaike information criterion. BIC = Bayesian information criterion. RMSEA = root mean square error of approximation. CFI = comparative fit index. SRMR = root mean square residual. Model 1 includes age and combat experience as covariates. Model 2 includes age, combat experience, and identity disruption as covariates.