 So the only way we can prevent ourselves from being fooled by inkblots is to make sure we have some way to establish that what we're seeing is a pattern and not a random occurrence. How do we fool ourselves with our children? Are we really seeing a change in behavior after we gave up dairy, gluten, soy, corn, apples, and chicken? Or are we seeing behavior because we want there to be an improvement? Did something else change in our lives to account for that change, like the passage of time, a new therapist, a change in routine, better sleep habits, etc? A few ways we bias ourselves:Bandwagoning: We tend to believe things are true because others around us think they are true or we hear the same thing all the time. Everyone thinks that chickenpox isn't dangerous and mercury causes autism. Research says the oposite. Confirmation Bias: We tend to think we're right about our chioces, so we'll look for information that supports it and ignore information that doesn't. Johny is doing so well today. It must be because we've taken chicken out of his diet. The fact that he melted down at dinner and spent half the evening stimming is just coincidence. He must just need more time off of chicken. This is related to halo effect - magnifying our conclusions about research or results based on our feelings about the people as individuals. Denial. Yuppers. If you don't want to accept that your child will have autism for the rest of their life, it's easy to accept alternative scenarios that allow for "recovery" and "cure." Although it's possible highly effective treatments or cures will exist in the future, we don't have such a thing now, no matter how much people want to believe. Framing: Context matters. If your focus on information is too narrow or too broad, you draw the wrong conclusions. I saw someone present a chart "proving" that vaccines didn't lower the incedence of infection. The chart covered too broad of a timeline and implied that the lower death rates from infection were due entirely to reasons other than vaccination. Yes, fewer people die of things than they did 100 years ago since we have improved treatment and sanitation, but if you look at what happens when people stop vaccinating, why those death rates go right back up. Illusion of control. Dude, we all want to feel like we're in control. So if it involves a magic incantation or a patent medicine, we'll have a tendency to see reasons why it works instead of seeing the reasons it doesn't. Mix in a bit of confirmation bias and stir. Post-Purchase rationalization: Related to confirmation bias, we rationalize that we made the right choice, even if we didn't. We're human. We make mistakes. It's ok to admit it. Cherry picking: related to confirmation bias, framing, and denial. If you point out a hand full of results that support your conclusion, you're not looking at the big picture. Anecdotes Are Not DataYou may have heard this before. You'll probably hear it again. Just because something seems to work for one child does not mean it works for every child. It could be a coincidence. It could be an unrelated event. The parents could just plain be fooling themselves. In order to know if something is working, we need to know if a statisticallly significant change occurs in a group. No, that doesn't mean that it works for everyone. Yes, statistics can account for changes that affect very small groups if the sample size is big enough. For most of the vaccine stuff, the sample size is more than big enough. Likewise, research has to be repeated to be believed. You can't go off what one guy says in one experiment and call it good. Just ask Dr Wakefield. Or don't ask him, because he'll tell yo he's right no matter what all the other experiments say. Yes, I ripped this list of references from Wikipedia. It's a good list, and while I wouldn't use Wikipedia as my sole source of information, it's not a bad starting point.
|