Skip to main content

View full-text article in PMC

. 2023 Aug 16;7:e43345. doi: 10.2196/43345

Table 2.

Model fit statistics for male data.

	Number of latent classes	BIC^a (LL^b)	AIC^c (LL)	AIC3 (LL)	Npar^d	Maximum BVR^e	Classification errors	Entropy R^2f
Model1	Cluster 1	30588.8978	30293.9984	30365.9984	72	202.584	0	1
Model2	Cluster 2	29487.7888	28877.5109	29026.5109	149	20.9173	0.0023	0.9871
Model3	Cluster 3	29521.672	28596.0156	28822.0156	226	24.6777	0.0021	0.9901
Model4	Cluster 4	29763.3869	28522.3521	28825.3521	303	22.1771	0.006	0.9778
Model5	Cluster 5	30114.1405	28557.7272	28937.7272	380	20.8918	0.0241	0.9365

^aBIC: Bayesian information criterion.

^bLL: log-likelihood (the smaller the absolute value of LL, the better the model fit).

^cAIC: Akaike information criterion.

^dNpar: number of estimated parameters.

^eBVR: bivariate residual.

^fEntropy R²: values >0.8 indicate high degree of separation between classes.