Is the logical error of concentrating on the people or things that made it past some selection process and overlooking those that did not.
Card 1 of 259. Answer: Is the logical error of concentrating on the people or things that made it past some selection process and overlooking those that did not.
About · Quizzes · History of AI · To suggest new questions or report an error send me a dm.
Survivorship bias is drawing conclusions from the cases that made it through some filter while ignoring the ones that didn't, which makes the survivors look more typical, or more successful, than they are.
The classic example is from World War II: the statistician Abraham Wald was asked where to add armour to bombers based on the bullet holes in planes that came back. He argued for armouring the places with no holes, because planes hit there were the ones that never returned. Everyday versions include studying only successful startups or funds that still exist, which makes risky strategies look safe.
In data science it shows up whenever the dataset is defined by an outcome: analysing only customers who are still active, training a credit model only on approved applicants, or measuring a feature's impact only on users who stuck around. The fix is to ask what process removed records from the data and, where possible, to include the failures.
