Sloppy statistics and research misconduct are nothing new, but it's rare to get a clear picture of exactly how questionable data gets turned into clicky headlines. We have that now with the latest reporting on Cornell food scientist Brian Wansink, and it's worth taking a minute to look at what's exactly so wrong about the dodgy research techniques he's been accused of employing.
Tagged With statistics
Misusing statistics is one of the most powerful ways to lie. Normally, we teach you how to avoid misinterpreting statistics, but knowing how numbers are manipulated can help you spot when it happens. To that end, we're going to show you how to make data say whatever the hell you want to back up any wrong idea you have.
Just how likely does "probably" sound to you? To some people, "probably" means that something is practically locked in. To others, it means the likelihood of something happening is highly dubious. This graph assigns percentage values to a range of common phrases relating to probability. Turns out you should say "almost certainly" instead of "probably" if you want to minimise doubt.
Scarcely a day passes when I don't receive a report from some analyst or research organisation informing me of how a new product has saved a bunch of companies a massive sum of money, or how a product has been identified as a leader or innovator in their chosen market niche. But can we trust these reports?
Making New Year's resolutions is tricky in the first place. Do you start out small with something you know you can achieve consistently without really pushing yourself out of your comfort zone, or do you go all out and hope that willpower alone will drag you through? In other words, do you make a resolution to catch the train to work a little more instead of driving, or do you make a resolution to take a holiday every month and pack on 20 kilos of muscle at the gym before the next year is out?
Housing affordability, high house prices and rents are attracting plenty of media attention right now. The latest figures on house prices, mortgages, number of first time buyers and so on are dissected by journalists and commentators as if this is an issue of recent origin. In fact what we have here is a long-term structural problem that has been neglected for decades.
Subscription streaming services like Netflix and Stan generally refuse to disclose how many hours of content they offer, but those numbers can be a useful raw metric for deciding who provides the best value. Streaming search app Gyde has toted up the numbers on the five main services in Australia right now. Who comes out on top?