The Uncounfoundedness Assumption

Although regression models are great for summarizing the association between variables, any causal interpretation of a regression models rests on the uncounfoundedness assumption (also known as the selection-on-observables assumption; and sometimes referred to as the conditional independence assumption). Which of the following summarizes this assumption?

Possible Answers

All variables that affect treatment assignment and the outcome of interest are observable and conditioned upon.
If something happens more frequently than expected during a given period, we can expect that it will happen less frequently in the future
There is little or no multicollinearity in the data
The residuals of a regression model are normally distributed.

Exercise

The Uncounfoundedness Assumption

Instructions

Possible Answers