AIC

ØAIC 

v AIC

AIC stands for Akaike Information Criterion (Akaike, 1987) and is used to measure the quality of the statistical model for the data sample used. AIC score useful only in comparison with other AIC scores of the same data set.

The lower the AIC value, the better.

AIC = 129,234

v BCC

BCC = Browne-Cudeck criterion used specifically to analyze moment structures and impose a larger penalty on less parsimonious models.

v BIS

BIC = Bayesian Information Criterion applies a larger penalty for complex models compared to AIC, BCC, CAIC and therefore has a greater tendency to select parsimonious models.

BIC = 234,567 for our tested (default) model.

v CAIC

CAIC = Consistent Akaike Information Criterion (Atilgan & Bozdogan, 1987) occurs only when means and intercepts are not explicit in the case of one group. CAIC applies a penalty to complex models that is higher than AIC and BCC, but less severe than BIC.

The reference values for these criteria depend on the sample size, number of parameters, and model complexity. As such, there are no fixed reference values. However, some guidelines suggest that models with lower AIC, BIC, BCC, and CAIC values are preferred over models with higher values.

Lower values of these criteria indicate better model fit.