Understanding Statistical Significance in Medicine
Author Information
Author(s): Westover M, Brandon K D, Bianchi M T
Primary Institution: Massachusetts General Hospital, Harvard Medical School
Hypothesis
Do physicians misunderstand the probabilistic logic of significance tests?
Conclusion
Most physicians misinterpret statistical significance, leading to incorrect conclusions about medical research.
Supporting Evidence
- Over 90% of physicians answered incorrectly on a quiz about statistical significance.
- Understanding Bayes' rule is essential for interpreting medical information.
- Significance testing is often misunderstood and misapplied in medical research.
Takeaway
Doctors often get confused by statistics, especially when it comes to understanding if a treatment works or not.
Methodology
A quiz was administered to 246 physicians to assess their understanding of statistical significance.
Potential Biases
Potential biases in self-reported understanding of statistics and the limited scope of the quiz.
Limitations
The study only included physicians from two major academic hospitals, which may not represent all physicians.
Participant Demographics
Physicians from two major academic hospitals.
Statistical Information
P-Value
<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website