Heterogeneity of borderline personality disorder symptoms in help-seeking adolescents

Table 3 Model fit comparisons for latent class analysis (LCA), confirmatory factor analysis (CFA), and factor mixture models (FMM).

Model	LL	No. of par.	BIC	SABIC	Entropy	BLRT
LCA
One-class	− 2797.12	9	5650.28	5621.71	1
Two-class	− 2542.70	19	5203.71	5143.40	0.74	≤.001***
Three-class	− 2508.00	29	5196.57	5104.52	0.66	≤.001***
Four-class	− 2488.52	39	5219.87	5096.09	0.63	≤.001***
CFA
One-factor	− 2518.92	18	5149.91	5092.77	N/A	N/A
Three-factor¹	− 2515.43	21	5161.62	5094.97	N/A	N/A
FMM
Two-class, one-factor
FMM-1	−2542.70	19	5203.71	5143.40	0.74
FMM-2	− 2517.43	21	5165.61	5098.96	0.08
FMM-3	− 2481.07	29	5142.70	5050.65	0.54
FMM-4	− 2472.06	37	5174.51	5057.07	0.55
Three-class, one-factor
FMM-1	− 2521.74	21	5174.23	5107.58	0.66	≤.001***
FMM-2	−2515.70	24	5180.83	5104.65	0.28	.08
FMM-3	−2472.28	40	5193.63	5066.66	0.53	.24
FMM-4	− 2450.36	56	5249.41	5071.66	0.60	.06

Notes. BIC Bayesian Information Criterion, BLRT parametric bootstrap likelihood ratio test, LL
Log-Likelihood, FFM factor mixture model, SABIC sample size adjusted BIC
¹ according to Sanislow et al. [24]

ISSN: 2051-6673