Further Reading: Survivorship Bias at Scale

Essential

Ellenberg, J. (2014). How Not to Be Wrong: The Power of Mathematical Thinking. Penguin. Contains an excellent chapter on survivorship bias, including the Wald story with mathematical detail. Accessible to non-mathematicians. (Tier 1)

Goldacre, B. (2012). Bad Pharma: How Drug Companies Mislead Doctors and Harm Patients. Fourth Estate. The most thorough and accessible treatment of publication bias in pharmaceutical research. Goldacre documents the structural mechanisms by which the file drawer distorts medical evidence. Essential reading for anyone in healthcare. (Tier 1)

Publication Bias

Turner, E. H. et al. (2008). "Selective Publication of Antidepressant Trials." NEJM, 358(3), 252–260. The landmark study comparing FDA data with published literature for antidepressants. The gold standard for publication bias documentation. (Tier 1)

Rosenthal, R. (1979). "The File Drawer Problem." Psychological Bulletin, 86(3), 638–641. The original formulation of the file drawer problem. Short, clear, and still cited after 45+ years. (Tier 1)

Ioannidis, J. P. A. (2005). "Why Most Published Research Findings Are False." PLoS Medicine, 2(8), e124. A landmark paper arguing that the combination of publication bias, low statistical power, and multiple comparisons means that a majority of published findings may be false. Controversial but influential. (Tier 1)

Business Survivorship Bias

Rosenzweig, P. (2007). The Halo Effect. Free Press. The best critique of the business success literature, including Good to Great and In Search of Excellence. Rosenzweig identifies the halo effect and survivorship bias as the central methodological flaws. (Tier 1)

Denrell, J. (2003). "Vicarious Learning, Undersampling of Failure, and the Myths of Management." Organization Science, 14(3), 227–243. A formal treatment of how survivorship bias in organizational learning creates persistent myths about management. (Tier 1)

The Wald Story

Mangel, M. & Samaniego, F. J. (1984). "Abraham Wald's Work on Aircraft Survivability." JASA, 79(386), 259–267. The academic treatment of Wald's contribution. (Tier 1)

Wald, A. (1943). "A Method of Estimating Plane Vulnerability Based on Damage of Survivors." SRG Memo No. 87. Declassified. Wald's original memo. Available online. More technical than the popular accounts but rewarding for mathematically inclined readers. (Tier 1)

Historical Survivorship Bias

Research on survivorship bias in historical evidence has been explored by historians working in subaltern studies (Guha), material culture studies (Schlereth), and the history of oral cultures (Ong, Goody). (Tier 2)

For Instructors

The Wald bomber example makes an outstanding classroom demonstration. Present students with the data (hit rates by section) and ask them where to put the armor. Most students will choose the fuselage. Revealing Wald's insight — and asking students to identify the error in their reasoning — creates a powerful "aha moment" that anchors the concept.