Correlation and Causation

I've said it here before. I'm sure I'll say it again. But for the edification of those who read this, repeat after me:

"Correlation is not Causation"

I first heard about this report on the radio this morning. As I read through the news, I ran across it again in print:

"American Youth TV Habits Lower Job Prospects, Community Service"

"Nov. 19 (Bloomberg) -- Americans aged 15 to 24 on average spend two hours a day watching TV and only seven minutes on leisure reading, reducing their chances for high-paying jobs and community service, according to a report by the National Endowment for the Arts."

"Sixty-one percent of those holding managerial or professional jobs were proficient readers, said the report, citing a 2003 U.S. Education Department survey. Some 70 percent of the people rated as poor readers felt their lack of skills had limited their job opportunities."

"The report concluded that 57 percent of those who had proficient reading skills had performed volunteer work, compared with 18 percent of the people with poor skills."

Reading less does not CAUSE people to volunteer less, or CAUSE them to not get high-paying jobs. Reading little does not, by itself, CAUSE someone to be a "poor reader." But we tend to mix up correlation (two things tending to move in the same direction, like volunteering and reading) and causation (one thing causing another thing to happen, like the gas tank going empty because you are driving). But by mixing up the two, a more interesting - though not actually valid - premise can be implied (or even stated outright), like "Not reading leads to bad citizens!"

The truth is that it's much more likely that something else causes both. The TV habits are a symptom not the problem. And as long as we only treat the symptoms, the problem only festers.