« Home | Photographer's image used in bid to discredit Kerry » | Weird » | No comment » | From the ER waiting room, part two: Saturday telev... » | Was waiting for this » | Er...? » | All things considered » | Bush good for something » | Lies, all lies » | Yes, our fault. Completely. So sorry. » 

Wednesday, February 18, 2004 

Repeat after me, 'correlation does not imply causation'

Violent video games have a much more damaging effect on children than parents would like to believe, leading them to perform poorly in school, argue with teachers, condone aggression and get into physical fights with their peers, according to a series of new studies.
- 'do video games breed violence?'; globeandmail.com; wednesday, february 18
At one point the reporter uses data that essentially says that higher levels of violent video game play tend to go along with lower levels of empathy to instead say, "greater exposure to violent video games causes lower levels of empathy and stronger pro-violence attitudes" (emphasis mine). This is false thinking. It may be true, but the type of data given after this statement does not give us enough to conclude that yet.

However, other than that point, I was impressed overall by the article.

Later the reporter writes:

"The final study shows that even brief exposure to a violent video game (Doom, in this case) can lead people to associate themselves with aggressive traits and actions. Half the students in the study played Doom for 10 minutes, while the other half played a non-violent puzzle game. They were then given a series of tests, designed to measure how aggressive they felt, and how they'd react to other people."

Because the study was experimental, we can actually say that the data from it suggests that violent video games may cause aggressive behaviour.

I hate it when reporters screw this up. Are they not given even the most basic training in science? In my first year of university there was one principle that was pounded into my head from every professor I had, whether they were sociology, psychology or philosophy. That principle was, "correlation does not imply causation." How hard would it be to teach reporters that?