Skip to main content

Advertisement

Table 3 Fit indices for models with different number of latent classes without covariates, for boys and girls separately

From: Gender-specific substance use patterns and associations with individual, family, peer, and school factors in 15-year-old Portuguese adolescents: a latent class regression analysis

Number of classes Free parameters LL AIC AICC BIC aBIC Entropy BLRT p-value
Girls
 1 29 − 4557.71 9173.43 9175.49 9311.81 9219.72 NA NA
 2 59 − 4026.9 8171.81 8180.51 8453.35 8265.98 0.836 < 0.001
 3 89 − 3883.03 7944.05 7967.51 8368.75 8086.11 0.809 < 0.001
 4 119 − 3800.78 7839.57 7877.49 8407.43 8029.51 0.807 < 0.001
 5 149 − 3739.76 7777.53 7839.35 8488.55 8015.36 0.835 < 0.001
 6 179 − 3699.53 7757.07 7850.06 8611.25 8042.78 0.859 < 0.001
 7 209 − 3668.88 7755.75 7888.15 8753.08 8089.35 0.866 < 0.001
Boys
 1 30 − 4052.14 8164.28 8167.16 8299.86 8204.61 NA NA
 2 61 − 3560.51 7243.03 7255.3 7518.69 7325.01 0.856 < 0.001
 3 92 − 3451.46 7086.93 7116.18 7502.69 7210.58 0.864 < 0.001
 4 123 − 3358.33 6962.67 7017.73 7518.52 7127.99 0.819 < 0.001
 5 154 − 3315.33 6938.66 7029.94 7634.6 7145.64 0.836 < 0.001
 6 185 − 3278.94 6927.88 7067.76 7763.93 7176.53 0.82 < 0.001
 7 216 − 3248.12 6928.24 7131.59 7904.38 7218.56 0.846 < 0.001
  1. Italic for best values
  2. LL log-likelihood, AIC Akaike information Criterion, AICC Corrected Akaike Information Criterion, BIC Bayesian Information Criterion, aBIC sample-size adjusted Bayesian Information Criterion, BLRT Bootstrapped Likelihood Ratio Test