Skip to main content

Advertisement

Table 6 Assessment of fit of categorical latent variable models—latent class analysis

From: Measuring behaviours for escaping from house fires: use of latent variable models to summarise multiple behaviours

  −2LL AIC SSA-BIC Entropy
1 Class −3139.53 6307.05 6332.76 1
2 Classes −2950.85 5959.70 6012.96 0.84
3 Classes −2875.93 5839.86 5920.68 0.77
4 Classes −2820.53 5759.07 5867.43 0.84
  1. −2LL minus twice the loglikelihood, smaller values indicate better fit
  2. AIC akaike information criterion, smaller values indicate better fit
  3. SSA BIC sample size adjusted bayesian information criterion, smaller values indicate better fit
  4. Entropy values close to 1 indicate more reliable classification (lower classification error)