3 ms·
The thing the article doesn't point out is why it seems unintuitive. If you phrased the question as "You have N pounds of potatoes", or with a specific number
by JoshTriplett 11y ago
The thing the article doesn't point out is why it seems unintuitive.
If you phrased the question as "You have N pounds of potatoes", or with a specific number other than 100, it would come across as less unintuitive. As you read, you see "100 lbs", and "99%", so percents and potato components are both out of 100. So then you see 98%, which is 98/100...
- vecter 11y agoAlthough it's somewhat true, if you had started with 14 pounds of potatoes instead, I still think most people at first glance would expect the final weight to be around 14, not half of it.
- JoshTriplett 11y agoI agree, but I suspect fewer people would make the mistake. Partly because more people would actually do the math, and partly because there's no immediately attractive wrong answer involving "98/100" and "100 lbs". There's still a somewhat attractive wrong answer of multiplying 98/100 by 14 lbs, but there's one less reason to make that mistake. This seems like the kind of thing that could be tested with a study. Ask a few hundred people each version of the problem (ideally filtering out anyone who has seen the problem before), and see how many get each version right. (Potato paradox problem paradox: You ask 100 people to solve the potato paradox. 99% of them get the answer wrong. You drop people who answered incorrectly until you have a group where 98% got the answer wrong. How many people are left?)
- DanBC 11y agoGerd Gigerenzer asks roughly similar questions of a wide range of people - mostly medical professionals. It's scary how wrong people are, especially considering that this is information they're using to medicate you or operate on you. Here's an example from his book: http://imgur.com/zO4zkl4 http://imgur.com/zO4zkl4 Gerd Gigerenzer Reckoning With Risk.