Individual items were grouped into constructs a priori. Single-group confirmatory factor analysis assessed each factor, with individual observer-rated items as measured indicators of each latent variable. Items with nonsignificant factor loadings were removed from the model. Fit was assessed using the χ2 goodness of fit index, the comparative fit index (CFI), and the standardized root mean square residual (SRMR) (Hu & Bentler, 1998). We used weighted root mean square residual (WRMR) instead of SRMR for models with categorical indicators (Hu & Bentler, 1999); weighted least-squares means-variance adjusted (WLSMV) (B. O. Muthén, du Toit, & Spisic, 1997) estimation; and DIFFTEST to estimate difference in χ2 statistics (L. K. Muthén & Muthén, 1998–2004). Criteria for assessing model fit included: (a) p value of χ2 (df) ≥ .05; (b) CFI ≥ .90; and (c) SRMR ≤ .08 or WRMR ≤ 1.00 (Hu & Bentler, 1999; McDonald & Ringo Ho, 2002; Vandenberg & Lance, 2000).
