What is it about?
When treatment effects are heterogeneous, the commonly used average treatment effect is only one metric of treatment outcomes. This article uses a Bayesian approach to estimate the distribution of treatment effects.
Featured Image
Photo by Goran Ivos on Unsplash
Why is it important?
The average treatment effect is an incomplete metric in settings where some may benefit and others may lose from treatment. This article provides a Bayesian approach to estimate the distribution of effects leading to a nuanced assessment of who benefits from treatment.
Perspectives
This was my first serious application of Bayesian analysis. Working with a long time colleague, Gerry Gaes, it was a challenge to work out the details.
William Rhodes
Read the Original
This page is a summary of: Estimating the Distribution of Treatment Effects From Random Design Experiments, Evaluation Review, February 2020, SAGE Publications,
DOI: 10.1177/0193841x20906232.
You can read the full text:
Contributors
The following have contributed to this page
