Skip to main content

Table 2 Fit statistics for 2- to 5-class solutions as derived from LCA at time points B1, B2 and B3

From: Co-occurrence, stability and manifestation of child and adolescent mental health problems: a latent transition analysis

Model

AIC

BIC

aBIC

Entropy

LMRT

p value

BLRT

p value

B1

 2-class

5.691.412

5.742.812

5.714.217

0.699

0.000

0.000

 3-class

5.675.873

5.755.828

5.711.348

0.862

0.000

0.000

 4-class

5.684.131

5.792.642

5.732.276

0.799

0.388

0.474

 5-class

5.694.131

5.831.197

5.754.946

0.671

0.232

1.000

B2

 2-class

5.488.616

5.539.821

5.511.226

0.681

0.000

0.000

 3-class

5.476.521

5.556.172

5.511.692

0.814

0.000

0.000

 4-class

5.486.494

5.594.592

5.534.226

0.700

0.545

0.978

 5-class

5.496.494

5.633.039

5.556.787

0.467

0.739

1.000

B3

 2-class

7.097.444

7.150.918

7.122.322

0.719

0.000

0.000

 3-class

7.077.793

7.160.976

7.116.493

0.855

0.002

0.000

 4-class

7.083.146

7.196.038

7.135.668

0.844

0.082

0.086

 5-class

7.093.146

7.235.746

7.159.490

0.579

1.000

1.000

  1. B1 = 2004–2007, B2 = 2005–2008, B3 = 2009–2012; AIC, Akaike information criteria; BIC, Bayesian information criteria; aBIC, adjusted BIC; LMR LTR p value, Lo–Mendell–Rubin likelihood ratio test p value; BLRT LRT p, Bootstrapped Likelihood ratio test p value; Best-fitting models by time points are indicated in bold