It shows how statistically extreme results are a predictable result of small sample sizes, and describes a common error -- the mistaken belief that any statistically extreme result requires a causal explanation.
Case studies discussed in the video include a discussion of kidney cancer rates in the United States, the Gates Foundation efforts to create better schools, and German bombing patterns on London during the "Blitz" (WWII).
People suck at reasoning with probabilities. We suffer from what some psychologists call "probability blindness". Our natural tendency to make judgments that either ignore or run contrary to basic laws of probability is one of the primary obstacles to critical thinking and rational decision making.
In this series I look at a variety of probability fallacies. The current set includes videos on critical thinking about coincidences and the Gambler's Fallacy, but I'll be adding additional videos on small sample fallacies, regression fallacies, base rate fallacies, and others.
In a later series of videos I'll be talking about the psychological origins of many of these fallacies, and what we can do to avoid falling into them or neutralize their negative effects.
Video versions of my podcast show on critical thinking. You can subscribe to the podcast on iTunes, there's both an audio and a video version. Just search for "Critical Thinker Podcast" in the iTunes search bar.
There is a small industry devoted to identifying and classifying fallacies of reasoning. A comprehensive list of recognized fallacies would run into the hundreds.
This course introduces the concept of a fallacy and discusses some common fallacy types, but it in no way aims to be comprehensive. Instead the focus is on how any given fallacy can be understood using the basic concepts of argument analysis introduced in earlier courses.
The only classification I use distinguishes logical or formal fallacies, fallacies that arise from false or implausible premises, and fallacies that are best understood as violations of one of the necessary conditions for having a rational argument at all.