Friday, May 7, 2010

anecdotes aren't evidence?

Humans go for fast thinking, not good thinking. If someone does something stupid in a crisis, people seem more likely to say "at least he did something" than, he should have stopped and thought that through.

This is shown by news stories complaining that the President chose to confer with his staff for very long periods of time before making a decision. Gathering of facts before stating something as true seems, perhaps, rather a waste of time.

But there is a huge difference between fact and opinion, between evidence and anecdote, between cause and effect and correlation... Most people don't even understand the terms, let alone how to distinguish them. Learning to think is not taught in schools. Regurgitating information in the same form it was given to you is what gets you good scores.

It's not just the way we are raised that determines this ability to think, either. The body, itself, is trained to respond to clues and come up with cause and effect. This is how one stays alive. If you eat a strawberry the same day you get the flu, your body is just as likely to grow allergic to the strawberry as any other reaction because it found a correlation.

But when this is the basis that we also approach science, we run in to problems.

We end up with a large group of parents saying that immunizations cause autism. We also end up with people finding correlations in nutrition that may not exist. We have doctors saying that fat causes heart attacks even though no study has fully proven this. All the studies have shown is that clots have fat in them. Never mind that clots aren't the biggest reason for heart attacks, nor type of fat (vegetable more often than meat), nor what causes clotting!

So most facts are, at best, incomplete. At worse, politically skewed.

Take the studies on whether or not carbs are harmful. The same studies are cited by quite a few different groups, all showing different results. And you can say that the corn industry backed a study that showed corn is healthy, and that the meat industry backed a study that said red meat is healing, and the PETA folk proved red meat causes cancer...

But when it comes down to it, what evidence of anything do we actually have? A handful of inconclusive studies (because they were not done well. Because nutrition studies are very difficult to do well), a lot of anecdotal evidence, and everyone calling "proof" wherever it fits their needs.

Now, I personally am a big fan of low carb. The studies I choose to buy into show that grains raise insulin and insulin is the basis of most modern epidemics (obesity, diabetes, auto immune disorders...). It is almost all based on anecdotal evidence (my friend went low carb and all her health issues cleared) and studies interpreted by low carb advocates. Do I have proof? No.

I'd love proof. I'd love to see some serious long term studies done. On most aspects of health and nutrition. Then again, I'd love for them to decide on what autism is, so it can be studied.

If you want to study a disease, you isolate the virus and you test it. You can immunize 5000 folk and not immunize another 5000 folk and study their blood to see who got which antibodies. You can discover patterns because there is a damned virus to look at.

No one knows what autism is. There are a handful of symptoms that people don't really agree on. There are behaviors that show in so many ways there's no real agreement there. They keep spreading the spectrum further and further out, making any real studies even more difficult. It isn't a disease. It's a behavioral pattern that annoys the piss out of most people. It is all the annoying traits most people have to a small degree... done huge and exaggerated.

Yet, people claim "proofs" and "evidence" and make diagnoses and cite studies and anecdotes and think they have some damned clue what is going on. The truth is, you don't have a definition to parse. Without a definition, there is nothing else. Form is beauty. Without form, there is nothing. Which is probably why people are so ugly about the disorder.

And that goes for all these things people claim "to know" without any real evidence.

Anecdotes are only that. Someone did something that seemed to have an effect. That's all. And that's perfectly fine to go by. There's nothing wrong with saying "my sister did this and it worked so I'm trying it". The problem is saying "and therefore this is true" or "therefore this is right". You don't have any kind of proof. You have a story of assumed correlation.

Bad science is very aggravating. Logical fallacies make me want to kick people in the head. What ever happened to "I statements"? They have a huge place in science. "I suspect" "I believe" "I have noticed"... What are people so scared of?


Jon A.S. said...

Hey Abi,

I agree that anecdotes are not proof or the truth...but are assumptions by people who, for lack of a better word, just make sh-t up to make a story or something,


Abifae said...

LOL. They are important, though. Noticing that every time we throw a ball in the air, it comes down... but that somehow birds stay up... All factors in our discovering information about gravity. Anecdotes lead people to investigate. It's only an issue when anecdotes are taken as proofs :)

OccamsTazer said...

This is a wonderful post, I love the way you think!
Bad science gives me all sorts of freak-outs.
I can't watch the news anymore for this reason...every story is sensationalized non-science.
"Is your microwave killing you? Find out at six!"
That makes me go AAAAGH!

Abifae said...

Thanks Occam!! I don't watch news at all. It's bad enough reading it and trying to find any kernel of truth through all the sensationalism.