Wednesday, April 14, 2010

The problem of scientific studies

Welcome back to my ongoing revival of posting.

Yesterday I shared with you the report of a new study on spanking. I ended that conversation with the question of a problem in logic. Here is the article again.

Study: Spanking Increases Aggression

The problem that I identified with the article was in the final paragraph.

Experts suggested that spanking might even be more dangerous to girls than boys because it trains them to be submissive. They said that can open women up to domestic abuse later in life.

The issue I saw is this:
- this article is about the study regarding an increase in aggressive behavior in children
- this article refers to children without any distinction between male or female becoming more aggressive due to spanking
- this article then claims that spanking is more dangerous to girls

Logic demands that the next statement should be: there is an overwhelming increase in dangerous aggressiveness among females. But is that where they go? Nope. Girls become submissive. On top of that, spanking opens women up to domestic abuse.

So is this a reporting problem or a study problem?

Here is something to consider. The Time article does not mention this "result". Neither does the Tulane University press release (where the study was conducted) about the study.

Now, in order to read the study you have to pay for it. And being the cheapskate that I am, that will not happen. But reading the abstract from Pediatrics shows nothing of the female specific increase in submissiveness. It also shows that the study focused on maternal spanking only.

So this comes back to bad reporting. And that is part of the problem of the relationship between scientific research and popular reporting. I say popular reporting because pure reporting is dry and uninteresting. If people read real reporting, they would read the study directly. Popular reporting tries to draw people into the issue they are reporting. And in general people lap it up. The general population does not bother to go and read the rest of the story.

If you read the news this is what you hear:

A new study supports scientifically that if you spank your child at all they will be aggressive, destructive children who are likely going to be bullies. And girls aren't children. They are a subspecies of creature that responds opposite of the findings.

Based on the abstract, here is what the study states:
Results: Frequent use of CP [corporal punishment, spanking] (ie, mother's use of spanking more than twice in the previous month) when the child was 3 years of age was associated with increased risk for higher levels of child aggression when the child was 5 years of age (adjusted odds ratio: 1.49 [95% confidence interval: 1.2–1.8]; P < .0001), even with controlling for the child's level of aggression at age 3 and the aforementioned potential confounding factors and key demographic features.

So what?

Before you react to something, find out the facts. If you read about a study that you don't agree with, go read it or get more information from different sources. Local news is not the only source. And the newspaper/television station/network of your choice will get the facts confused sometimes.