Statisticians can prove almost anything, a new study finds | News | National Post | Science News | Scoop.it

In “False-Positive Psychology: Undisclosed Flexibility in Data Collection and Analysis Allows Presenting Anything as Significant,” two scientists from the Wharton School of Business at the University of Pennsylvania, and a colleague from Berkeley, argue that modern academic psychologists have so much flexibility with numbers that they can literally prove anything.