The role of causal reasoning in understanding Simpson's paradox, Lord's paradox, and the suppression effect: covariate selection in the analysis of observational studies
2008
Understanding Statistical Paradoxes in Causal Analysis
Commentary
Author Information
Author(s): Arah Onyebuchi
Primary Institution: Department of Social Medicine, Academic Medical Center, University of Amsterdam
Conclusion
The study emphasizes that statistical criteria are inadequate for causal analysis and that causal reasoning is essential.
Supporting Evidence
- The paper discusses the equivalence of Simpson's paradox, Lord's paradox, and the suppression effect.
- It highlights the importance of covariate selection in causal analysis.
- The author argues that causal reasoning is necessary to understand these statistical phenomena.
Takeaway
Sometimes, statistics can be misleading, and we need to think about the real causes behind the numbers instead of just the numbers themselves.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website