Hi and welcome!

This website gathers code and additional material for the paper Causal Exaggeration: Unconfounded but Inflated Causal Estimates.

This paper shows that the use of causal identification strategies induces a trade-off between two biases: avoiding confounding and exaggerating true effect sizes due to a loss in statistical power.

This paper introduces the ididvar package. More information on the package is available on its website.

Resources: Paper , Non-technical summary