Table 5. Goodness-of-fit statistics.

	np	χ2	p-value	AIC
Full model	24	19.325	0.003	122.462
Identifiable model	6	15.473	0.906	82.61

In the full model, 24 parameters were estimated. After identifiability analysis, estimated parameters were reduced to 6 and the remaining parameters were fixed prior to fitting. The reduction in estimated parameters improved the weighted least squares merit function value (χ²), increased p-value on a χ² test indicating that the identifiable model sufficiently explains the data, and lowered the estimated amount of information lost between the model and the data by the Aikake Information Criterion (AIC) measure.