Confounding and collapsibility in causal inference

Consideration of confounding is fundamental to the design and analysis of s tudies of causal effects. Yet, apart from confounding in experimental desig ns, the topic is given little or no discussion in most statistics texts. We here provide an overview of confounding and related concepts based on a co unterfactual model for causation. Special attention is given to definitions of confounding, problems in control of confounding, the relation of confou nding to exchangeability and collapsibility, and the importance of distingu ishing confounding from noncollapsibility.