Skip to main content
. Author manuscript; available in PMC: 2011 May 1.
Published in final edited form as: Compr Psychiatry. 2009 Aug 27;51(3):224–235. doi: 10.1016/j.comppsych.2009.07.002

Table 2.

Fit Indices for Dual Trajectory Model Solutions

Classes AIC BIC Entropy
2 9012.098 9060.791 .828
3 8708.124 8783.869 .768
4 8559.691 8662.488 .733
5a 8443.005 8572.853 .753
6a 8381.070 8537.970 .728
7a 8364.654 8548.606 .742
8a 8344.484 8555.487 .730
9a 8327.887 8565.942 .738
10a 8317.993 8583.100 .760

Note: AIC = Akaike Information Criteria; BIC = Bayesian Information Criteria.

a

Beginning with the 5 class solution, constraints on the threshold and slope parameters anxiety disorders in the class reflecting a consistently very low probability of having an anxiety disorder were necessary for appropriate model estimation. Lower BIC values indicate better fit. Higher entropy values indicate greater precision of class membership assignment.