When people are told a statistical claim, particularly about risk, the most important follow-up they can ask is about magnitude. How big of a number or impact is this? But many lack a basic understanding of statistics and how they fit into our world. It’s not baked into the fundamentals of education.
David Spiegelhalter is an emeritus professor of statistics at the University of Cambridge. His books like The Art of Statistics: How to Learn from Data and Covid By Numbers: Making Sense of the Pandemic with Data help contextualize the importance and impact of statistics in everyday life.
He and Greg discuss the vital role of data literacy, how concepts like 'micro-mort' and 'micro-life' can measure risk, and the ramifications of faulty statistical interpretation during crises like COVID-19.
*unSILOed Podcast is produced by University FM.*
Episode Quotes:
You can't talk about risk without talking about its magnitude
04:31: To talk about risk without talking about magnitude, I think, is an abominable thing to do. It's manipulative. It's always manipulative. If someone's going to talk about risk, they are trying to worry you, and they're trying to manipulate your emotions. Most of the time, we talk about increased risk and the risk (delete the ‘of” )without having any idea what the magnitude is. And even if we do, it's quite difficult to know: is that a big number or not? So, I think that this is absolutely essential: whenever people are told something, a claim, they should ask, "How big is it? And is this really a big number? Is this really important?"
Risk as analysis is very often dominated by risk as feeling
21:22: Risk as analysis is very often dominated by risk as feeling, and you've got to have risk as feeling, I think, in there as well. But it's when one of them takes over. And I think the real problem with this is that if you just operate on risk as a feeling, it's so easy to be manipulated by people. "Oh, this is awful. This is awful." You've got to be really careful of this. And you think, "No, it's not bad," or understating what the risks of some things are. So I think that if you're very vulnerable, if you only operate on risk as a feeling, you're vulnerable to manipulation.
How do you gain micro life?
38:40: You can gain micro-lives by exercising and stuff like that. And so, but that's highly nonlinear. The benefits from the first 20 minutes of exercising are considerable. It's about 40 minutes. So it's about a micro-life and a half, or something like that. After that, it's about the past. So, if you exercise for half an hour, you live an extra half an hour. So you better enjoy exercising because that's the extra bit you're living. And it's like time, and I quite like this image that while you're exercising moderately, your aging stops. You're not aging that half hour.
COVID's positive impact on public interest in data
47:00: During COVID, it was amazing. The popular interest in statistics, data, and graphs—I was on the media all the time trying to explain stuff. And that's carried on. It's even good. Is there any good news about COVID? One of the small things I think might be the greater public tolerance for an interest in data and graphs and more subtle ideas being used.
Show Links:
Recommended Resources:
Guest Profile:
His Work:
unSILOed with Greg LaBlanc 