1. From: Toward Evidence-Based Medical Statistics. 1: The P Value Fallacy
Ann Intern Med. 1999;130(12): doi: /
Figure Legend:
The parallels between the processes of induction and deduction in medical inference (top) and statistical inference (bottom).Δ = treatment difference.
Date of download: 11/14/2017
Copyright © American College of Physicians. All rights reserved.

2. From: Toward Evidence-Based Medical Statistics. 1: The P Value Fallacy
Ann Intern Med. 1999;130(12): doi: /
Figure Legend:
Bayes theorem, in words.
Date of download: 11/14/2017
Copyright © American College of Physicians. All rights reserved.

3. From: Toward Evidence-Based Medical Statistics. 1: The P Value Fallacy
Ann Intern Med. 1999;130(12): doi: /
Figure Legend:
The bell-shaped curve represents the probability of every possible outcome under the null hypothesis.PPBoth α (the type I error rate) and the value are “tail areas” under this curve. The tail area for α is set before the experiment, and a result can fall anywhere within it. The value tail area is known only after a result is observed, and, by definition, the result will always lie on the border of that area.
Date of download: 11/14/2017
Copyright © American College of Physicians. All rights reserved.

4. From: Toward Evidence-Based Medical Statistics. 1: The P Value Fallacy
Ann Intern Med. 1999;130(12): doi: /
Figure Legend:
Possible outcomes of two hypothetical trials in six patients (Appendix). The only possible overlapping results are the observed data and the result in which treatment A was preferred in all patients.
Date of download: 11/14/2017
Copyright © American College of Physicians. All rights reserved.