Modelling latent variables

Latent variables are variables that are not directly observed but are theoretically or empirically derived from observed variables. To use these latent variables we use methods such as the following:

  • Structural equation models
  • Multilevel models
  • Post-response models
  • Ordinal logistic regression
Latent variables are derived from variables that have been observed.

The research line Modelling latent variables has four sub-areas:

  • Item response theory
  • Structural equation modeling
  • Multilevel modeling
  • Logistic prediction model