Abstract
Three related scenarios are presented for which the probability of an event appears to be larger than each of its mutually exclusive and exhaustive conditional probabilities. While the scenarios are structurally different and each has its own specific explanation, they illustrate the importance of an underlying principle that is sometimes overlooked in statistical analysis.
Additional information
Notes on contributors
Milton W. Loyer
Milton Loyer is the statistician for the Penn State Fruit Research and Extension Center in Biglerville, Pennsylvania, and an adjunct professor of mathematics and statistics at Lycoming College in Williamsport, Pennsylvania. He earned his PhD in statistics from Montana State University in Bozeman, Montana, and has taught full-time at undergraduate institutions in North Carolina, Arizona, and Pennsylvania. He is a member of the Harrisburg, Pennsylvania, chapter of the ASA and is a two-time winner of the award for best contributed statistical education presentation at the Joint Statistical Meetings.