Skip to main content
. Author manuscript; available in PMC: 2022 May 1.
Published in final edited form as: Brain Behav Immun. 2020 Sep 2;94:369–380. doi: 10.1016/j.bbi.2020.08.030

Table 2.

Fit Statistics and Statistical Parameters for Four Latent Class Models: a One Class Model, a Two Class Model, a Three Class Model, and a Four Class Model (n = 1,066)

Fit Statistics Model 1: One Class Model Model 2: Two Class Model Model 3: Tdree Class Model Model 4: Four Class Model1
Log Likelihood −333.79 −266.10 −221.14 −199.44
AIC 683.59 554.20 470.28 432.89
BIC 723.36 608.89 539.89 517.41
Entropy 0.76 .77 .79
LRT p < .001 p < .001 p = .11
Model 1: One Class Model Model 2: Two Class Model Model 3: Tdree Class Model Model 4: Four Class Model
 Classes Intercept Slope N Intercept Slope N Intercept Slope N Intercept Slope N
Class One .35** .01* 1066 .32** −.03* 883 .25** .01 800 .24** .01 783
Class Two Not estimated – one class model .49** .17* 183 .47** .21** 136 .28** .30** 75
Class Tdree Not estimated – two class model .81** −.21** 130 .79** −.25** 114
Class Four Not estimated – tdree class model .71** .06 94

AIC = Akaike information criterion; BIC = Bayesian information criterion; LRT = adjusted Vuong-Lo-Mendell-Rubin Likelihood Ratio Test(p value)

*

= p <.05

**

= p <.001

1

= Failed to reliably converge on a single solution.