This content is locked. Please login or become a member.
Perverse Incentives
My book, May Contain Lies, is about misinformation, but with a twist. We often think that the solution to misinformation is to check the facts. But what I point out is just checking the facts is not enough. So even if a fact is a hundred percent accurate, it could still be misleading. It could be a single anecdote, the exception that doesn’t prove the rule, or it could be a large-scale correlation when there’s no causation.
So what prompted me to write the book is the huge amount of misinformation out there. So it’s produced from a huge range of fields, not just from politicians and journalists, but also authors and even respected scientists. Some of the leading academics have been writing papers which end up being retracted. How is it that leading academic scientists who are well-trained in research can produce misinformation? Well the reason is that scientists are people too, so they’re subject to the same incentives as the rest of us.
So, often if you’re doing research it may well be that if the research gives a message that people want to be true, it’s likely to go viral. It’ll be shared, and tweeted, and blogged about, and it can reach millions. And even if reality is nuanced, if things are not black and white, if you give a black-and-white message, it’s easier to remember. It’s easier to catch on. As researchers, historically, scientists’ only ambition was to publish in the top academic journals. But now, if you’re a leading researcher, you could write a best-selling book, you can have a TED talk, and so there can be incentives to simplify your message or slightly distort it to give messages that people want to hear.
My biggest concern with misinformation is that people will claim that something is rigorous because it’s by an authority figure or it’s written in a book. So people might say, well, there’s a book which shows this, but anybody can write a book. There is no quality control. There is no peer review of most books.
A Case Study in Health Advice
A major domain in which there is a lot of misinformation is health advice because people genuinely want to improve their health or get fit or improve habits. And so if you give a simple piece of advice, which is black and white and easy to implement, then it might go viral even if it’s not backed up by data.
And one example is the Atkins diet. So the Atkins diet recommended eating as few carbs as possible. He said minimize all carbs period. Not just simple carbs, and saying that complex carbs are fine — all carbs. And that is simple to implement because you only need to look at the nutrition information label on an item of food to see the level of carbs. You don’t need to worry: Are these carbs simple or complex? Just avoid anything with high carbs. And also, he said, have as few carbs as possible, not that carbs are fine as long as they are thirty to fifty percent of your daily calories. That would be hopelessly complicated to implement. You would need to track all of your calories from carbs, and fats, and proteins, and then look, are my carbs within this small range?
So a simple rule, a black-and-white rule such as avoid all carbs — that’s something that can easily catch on even if it’s not backed up by science. In fact, doctor Atkins’ Diet Revolution, that’s the book that spawned the Atkins Diet, the publishers made him take out references to scientific research because they said, well, this is not a book which will be read by scientists. This is a book that will be read by a general audience. And so the simpler you give the message, the more likely it is to sell. Atkins didn’t need to be right. He just needed to be extreme.