Why Most Published Research Findings Are False A Research Paper by John P.A. Ioannidis journals.plos.org There is increasing concern that most current published research findings are false. The probability that a research claim is true may depend on study power and bias, the number of other studies on the same question, and, importantly, the ratio of true to no relationships among the relationships probed in each scientific field. In this framework, a research finding is less likely to be true when the studies conducted in a field are smaller; when effect sizes are smaller; when there is a greater number and lesser preselection of tested relationships; where there is greater flexibility in designs, definitions, outcomes, and analytical modes; when there is greater financial and other interest and prejudice; and when more teams are involved in a scientific field in chase of statistical significance. researchsciencetruth
The Wile E. Coyote Effect I’ve been looking at this chart a lot over the past few weeks. It shows us that print ad budgets were doing just fine all the way though the first decade or more of the consumer internet. There was even a little spike upward for the Dotcom bubble. Then the financial crisis and recession of 2008/9 caused a step change down, but when the crisis was over the budgets didn’t come back. Instead, the market had been reset, and budgets have been falling steadily ever since. You might call this the Will E Coyote effect - you’ve run off the cliff, or the cliff has disappeared from under you, but there’s a brief moment while your legs windmill in the air before gravity kicks in. It can take a while for the inevitable to happen, but then, as Lenin pointed out, you get a decade of inevitable in a week. Benedict Evans, COVID and cascading collapses www.ben-evans.com When decades happen business