Slashdot Mirror
Candy Linked To Violence In Study
T Murphy writes "A study published in the British Journal of Psychiatry links daily consumption of candy at the age of 10 to an increased chance of being convicted of a violent crime by age 34. The researchers theorize the correlation comes from the way candy is given rather than the candy itself. Candy frequently given as a short-term reward can encourage impulsive behavior, which can more likely lead to violence. An alternative explanation offered by the American Dietetic Association is that the candy indicates poor diet, which hinders brain development. The scientists stress they don't imply candy should be removed from a child's diet, although they do recommend moderation. The study controls for teachers' reports of aggression and impulsivity at age 10, the child's gender, and parenting style. The study can be found here, but the full text is behind a paywall."
While I agree that a sample of 35 isn't great statistics, the odds of having 69% of them in the candy-eating category if they WERE the same as the background population is under 0.05%, as I'm sure you know. (I just did a Monte Carlo simulation with 100000 trials.) So it's not the best study in the universe, but this is real human data: you take what you can get, particularly in sample size. It's not enough of a study to drive policy, but it's certainly enough to be publishable and enough to warrant further attention.