Interpersonal predictors of loneliness in Japanese children: variable- and person-centered approaches

BMC Psychology

Table 3 The summaries of model comparison of latent profile models at both T0 and T2

Model	df	AIC	BIC	aBIC	∆LR	p
T0: Baseline
One-profile model	10	12,103.91	12,153.72	12,121.96
Two-profile model	16	9253.27	9332.96	9282.14	2862.65	< .001
Three-profile model	22	7741.47	7851.34	7781.47	1523.51	< .001
T2
One-profile model	10	12,155.41	12,205.17	12,173.41
Two-profile model	16	8615.42	8695.04	8644.22	3551.99	< .001
Three-profile model	22	7141.72	7251.20	7181.33	1485.69	< .001

AIC = Akaike’s information criterion. BIC = Bayesian information criterion. aBIC = sample size adjusted BIC. ∆LR means difference of likelihood ratio between neighboring class models (i.e., comparing k − 1 and the k class models)

ISSN: 2050-7283