The Covariance Structure of Depression

There is a relationship between the underlying latent variable and the measured variables that goes in a very specific direction, right? Yes. And so we're moving into then causal indicators, which I think is a wonderfully fun topic to argue about. There have been years and years and years of published research that use multiple indicator latent factors for substance use. It is a fundamentally causal statement that the reason you observe the covariance structure among the items is because they share an underlying causal influence.