Non-Experimental / Observational Data

Non-Experimental/Observational study are the ones where one observes the effect of the treatment in a group of people without trying to make any intervention or changes. We cannot control for the treatment assignment in observational data. The Population with elements that have the same characteristics are compared with respect to the exposure to make Causal Conclusions.

Randomization is not always possible( and unethical sometimes), which is why most of the data is in the form of observational data. Making conclusions from Observational data is really hard. It is in fact one of the biggest problems in causality.

Note: Experimental data is too expensive and also illegal sometimes, which is why we heavily depend on Observational data for the causal conclusions. This is the part of Causality, where a lot of research and innovation is being done.

****