Skip to main content
. Author manuscript; available in PMC: 2018 May 1.
Published in final edited form as: J Youth Adolesc. 2016 Nov 9;46(5):1057–1075. doi: 10.1007/s10964-016-0596-x

Table 3. Fit Statistics for Successive Latent Class Models.

Model AIC Adjusted
BIC
BIC BLRT
p value
LMR LRT
p value
Entropy Error
Message
Without Covariate
One Class 2769.95 2774.06 2786.76 - - - no
Two Class 2443.51 2450.71 2472.92 <.001 <.001 0.78 no
Three Class 2306.83 2317.11 2348.85 <.001 <.001 0.90 no
Four Class 2291.95 2395.32 2346.58 <.001 <.01 0.84 no
Five Class 2275.31 2291.76 2342.55 <.001 0.44 0.84 no
Six Class 2212.11 2231.65 2291.96 <.001 <.001 0.88 no
Seven Class -41.47 -18.84 50.99 <.001 .91 0.99 yes

With Covariate
Five Class 2244.06 2263.60 2323.90 <.001 0.23 0.85 no
Six Class 2185.30 2207.93 2277.76 <.001 <.05 0.87 no
Seven Class 2165.64 2191.35 2270.70 <.001 0.10 0.88 yes

Note. AIC = Akaike Information Criterion; BIC = Bayesian Information Criterion; BLRT = Bootstrapped Likelihood Ratio Test; LMR LRT = Lo–Mendell–Rubin Likelihood Ratio Test. Child sex was subsequently included as a covariate with direct effects on the indicators, i.e., Perceived Mom and Dad Achievement Emphases, improving the overall fit of the model.