Pyramid Power: Searching for an Error in Test Scoring with 830,000 Helpers

A mathematics question that was asked on the Preliminary Scholastic Aptitude Test was scored incorrectly.This subsequently was discovered and became the subject of national attention. In this article we examine the data generated by this item from almost 830,000 examinees and find that detailed statistical analysis with even this enormous sample size would not have yielded clues to the blunder.The epistemological and practical consequences of this are also discussed.