Next time you read a study about why you should eat a certain food to stay healthy, keep in mind this entertaining study showing why that first study is probably bogus. "You Can't Trust What You Read About Nutrition" is the headline of the post from the stats gurus at FiveThirtyEight.com. The analysis by Christie Aschwanden zeroes in on a fundamental flaw of nearly all such stories—they're based on participants' self-reporting. These food diaries and food frequency questionnaires (or FFQs in researcher jargon) are notoriously unreliable for a host of reasons—faulty memory, embarrassment, confusion over portion sizes and ingredients in meals, you name it. Aschwanden can attest to this firsthand, having submitted to a six-month FFQ administered by a professional. Two co-workers and reader volunteers also participated.
Based on these results, it seems that Aschwanden consumes twice the calories of a female colleague, even though they are the same height and weight. Which seems ... off. But the best part is when the team puts its stats expertise to work to find all kinds of associations from their surveys, in the same manner that nutrition studies do. They found links between cabbage and innie bellybuttons, raw tomatoes and Judaism, and potato chips and higher math than verbal scores on the SATs. "Our experiment found that people who trim the fat from their steaks were more likely to be atheists than those who ate the fat that god had provided for them," writes Aschwanden. It was "silly easy" to find such correlations, she adds, which should give people pause when headlines about the next superfood come rolling around. Click for the full post.