Found a new economics blog with an entry today about the difference between causation and correlation. The author writes of an advertisement she saw from a food producer. They claimed that kids from families that eat dinner together get better grades in school. Of course there are lots of reasons some kids do better in school, and some of those reasons might also be correlated with eating dinner with the family.
This correlation-causation mistake happens commonly, like an advertisement I see regularly when checking my email. They say that kids who try alcohol before age 15 are three times more likely to become problem drinkers as adults. They want you to talk to your kids about drinking before this happens. Of course there are lots of reasons for problem drinking that might also be related to the likelihood of a kid drinking at a young age–not much supervision, over-active sense of adventure or curiosity, an unattractive personality.
What? You gonna pull your ten year old aside and tell em not to drink? Naaah. Better to have dinner with them.