Skip to main content
. Author manuscript; available in PMC: 2015 Jun 1.
Published in final edited form as: Behav Res Methods. 2014 Jun;46(2):488–498. doi: 10.3758/s13428-013-0391-5

Table 3.

Tests of Longitudinal invariance of measurement model of RSES with longitudinal CFA method.

Model χ² df RMSEA CFI TLI SRMR Δχ² Δdf p ΔRMSEA ΔCFI
1. Configural invariance 402.7 137 .026 .982 .975 .032
Model 1 versus Model 2.a. 70.1 10 <.0001 .002 .004
2.a. Metric invariance of global(GSE) factor* 473.1 147 .028 .978 .971 .040
Model 2.a. versus Model 2.b. 350.5 10 <.0001 .009 .022
2.b. Metric invariance of both positive and negative method factors** 791.0 157 .037 .956 .947 .043
Model 2.b. versus Model 3 210.9 10 <.0001 .003 .009
3. Scalar invariance 951.1 167 .040 .947 .939 .046

Note: RMSEA = root mean squared error of approximation; CFI = comparative fit index; TLI = Tucker-Lewis index; SRMSR = standardized root mean squared residual. Δχ²= Satorra-Bentler scaled (S-B scaled) χ²difference test. The latent variables were identified with fixing one factor loading being equal with 1.

*

The factor loadings on global factors are constrained to be equal in Time 1 and Time2.

**

The factor loadings of both method factors are constrained to be equal in Time 1 and Time 2.