Results from MGCFA suggested the measurement model allowing factor loadings to be freely estimated across racial groups (χ2 = 31.66, df = 10, CFI = .986) and the model constraining factor loadings to equality across racial groups (χ2 = 33.91, df = 14, CFI = .987) were not significantly different in terms of model fit (Δχ2= 2.25, Δdf = 4, p = .69, ΔCFI = .001), suggesting that factor loadings were invariant (i.e., metric invariance) across the EA and AA groups in the measurement model assessing fathers’ and mothers’ positive parenting. The Wald test for simultaneously constraining all of the key path coefficients in the model to equality across ethnic groups was not significant (Δχ2 = 14.69, Δdf = 12, p = .26), suggesting that there were no ethnic differences in these path coefficients.
