This article argues that nutrition studies are often unreliable because they rely on inaccurate self-reported data and can be easily manipulated. via FiveThirtyEight