Latent variables and Causal inference
When Biological datasets are analyzed it’s very important to understand that there are two types of variable constructs to keep in mind. Variables that are in the dataset, the variables that can be measured, and turned into numerical data are called the known or observed variables. We may include these in the models, perform statistical inference and other statistical methods. But there are always hidden, latent variable constructs which most often affect the results. These variables belong to the latent variable construct and are called the latent variables. They are important because most of the time in biological experiments and studies there are hundreds if not thousands of unobserved, latent variables which influence the results. They often introduce the bias and increase statistical errors in analyses. They also contribute to the variation in the data, mostly increasing the variation.
The question is what can we do about them? Well...
