James A. Smith: The method that can "prove" almost anything

The method that can "prove" almost anything

In 2011, a group of researchers conducted a study designed to find an impossible result. Their study involved real people, truthfully reported data, and commonplace statistical analyses. So how did they do it? The answer lies in a statistical method scientists often use to try to figure out whether their results mean something, or if they're random noise. James A. Smith explores p-hacking. [Directed by Anton Bogaty, narrated by Addison Anderson, music by Salil Bhayani, cAMP Studio].

Statistical tricks can make almost any research result look convincing, even the claim that a song makes people younger. By selectively choosing which data and analyses to publish, researchers can produce statistically significant results that are actually meaningless. The method known as p-hacking exploits the confusion around p-values, numbers commonly trusted in science to separate signal from noise. Pre-registering experiments and making scientific practices more transparent can help safeguard against these misleading outcomes.