Wednesday, January 14, 2009

The Dingbat Effect: What happens when twits use statistics they do not understand

Psychological research, like most social science research, is really difficult to conduct because what you are measuring, testing, and/or comparing doesn't really exist in an objective, material sense. I'm talking about thoughts, emotions, and beliefs. There are some things we can measure like behaviors, rates of stuff, and correlations between things. The reason why it is so difficult to get a PhD in the social sciences is that you really need to prove your ability to independently define, measure, and analyze abstract phenomena and know how to interpret numbers into meaningful bits information.

I'm proud of the work I do and I know that I have the skills to interpret statistics in substantive ways. The bad rap that social science research gets usually comes from those instances when people, who often do not have the training in statistical analysis, take pieces of data and interpret them however the hell they want to argue a point. Figure A: Ann Coulter—classic dingbat who doesn't know how to interpret data or statistical trends. She has been popping up across the media outlets lately to promote her latest hate-filled tome (I won’t say the name here so I can make it through this post without getting sick). She essentially blames the ills of society on single mothers because "forty years of social science research" has shown that the majority of individuals in jail, namely male inmates, come from single parent homes. There are so many things wrong with this interpretation, let me outline a few here: (ugh, where to start?)

1. Though the correlation may be true, that most inmates come from single parent homes, this remains a correlation. This does not mean, as Coulter likes to frame it, that single parent homes more often than not create criminals. Correlation does not mean causation.

2. Single parenthood is correlated with socioeconomic level. Coulter fails to acknowledge this HUGE detail. Socioeconomic barriers have been shown to be associated with crime rates. This is an issue of co-variability. Had Coulter not ignored this issue, she may have sung a different tune. You cannot just ignore the fact that financial hardship is a factor that is correlated with family structure.

3. Aren’t you forgetting a large segment of the population? What about individuals from single parent homes who are not in jail? What about the soldiers that bravely defend our nation? How many of them come from single parent homes?

4. Having two parents does not guarantee anything. Quality of family overrides quantity. Growing up with an abusive parent is associated with more negative psychosocial consequences than not having a parent around. Similarly, number of parents does not guarantee that parenting is actually going on. This is also a culturally biased view as many non--Euro/American individuals define their families as their nuclear unit plus the support of extended families. If your dad isn't around but you have 10 aunts, uncles, and cousins looking out for you, are you guaranteed a jail sentence?

5. Coulter assumes her moral perspective is the right one. This perspective, being the socially conservative anti-sex education, anti-social services, anti-choice, guarantees nothing when it comes to teen pregnancy and crime. Teen pregnancy is higher in politically and socially conservative communities than in liberal ones. Guess who's having babies without a partner, Ann? Do you social science homework. Telling kids to not have sex doesn't work.

6. What is her analysis of two-parent homes and success in child rearing? How would she define successful childrearing? Given the numbers she alludes to, it would seem that any trends she notes wash out when we take socioeconomic level into consideration, not just race. If she thinks that just taking into account race is rigorous analysis, than she is racist. No surprise there.

7. What do you know of the fathers that don't hold up their end of things? Perhaps she should look into why single mothers are single. Are they widowed? Were they abused? Is the father of their child in jail? Hmmm.

8. Using the jail system as a basis for identifying what's wrong with our country is biased and unsound. Due to the unfortunately realities of racial profiling, which is alive and well today, you can't really generalize anything from the prison population. How was the data that Coulter so eagerly references collected? Did it only include inmates who speak English and can read? Did it include female inmates? There are too many questions to list that would undermine Coulter's sad little point.

One of the worst crimes you commit in research is failing to consider the other side—the people who you aren't examining. Even worse is acknowledging contrary evidence to your point and simply ignoring it, instead of addressing it with thoughtful discourse. Coulter is not only stupid, she's rude and doesn't know how to debate.

The saddest part of Coulter's pathetic diagnosis of our nation's presenting problems is that she and the majority of political conservatives have been against the very programs that could help single parents level the playing field. They need support, yet the conservatives would rather punish these families by abandoning them under a perverse application of evolutionary survival of the fittest. If a woman can raise kids on her own, I would say that is pretty damn fit!

I'll be hopping off my soapbox now. I just needed some healthy venting. Ahh.

No comments: