Group items tagged

By UTS' Stephen Woodcock. Statistics is a useful tool for understanding the patterns in the world around us. But our intuition often lets us down when it comes to interpreting those patterns. In this series we look at some of the common mistakes we make and how to avoid them when thinking about statistics, probability and risk.

Why probability and understanding statistical fallacies matters Some of our most important decisions are shaped by something as random as the order in which we make them. The gambler's fallacy, as it's known, affects loan officers, federal judges -- and probably you too. How to avoid it? The first step is to admit just how fallible we all are.

"Fenofibrate may reduce heart disease risk in some patients with type 2 diabetes" The release goes on to suggest that among patients with type 2 diabetes who also had high levels of triglycerides and low levels of "good" cholesterol, the drug lowered the risk of cardiovascular events compared with placebo. Sounds like it might be time to ask your doctor about the benefits of fenofibrate, right? Before you schedule that appointment, you might want to consider this detail that wan't included in the news release: The findings of benefit came from a small group of patients within a larger study whose results had previously been reported. And that larger study, known as ACCORD, found no overall cardiovascular benefit among patients treated with fenofibrate.

Not a day goes by without a flurry of data-driven memes passing through my Facebook news feed, sailing by on my Twitter stream or landing as alerts in my email inbox that cite what appear to be reputable datasets and using them to offer surprising conclusions, typically wrapped up in a mesmerizing infographic. Yet, when I pick any of these memes at random and delve into it, I find that it is the rare meme indeed that stands up to statistical scrutiny.

Excellent analysis from the FT (you'll need to login to view via the link) that uses knowledge of the Mean and Median to show that some companies have reported incorrect (fabricated?) pay-gap information! One in 20 UK companies that have submitted gender pay gap data to the government have reported numbers that are statistically improbable and therefore almost certainly inaccurate, a Financial Times analysis has found. Sixteen companies, each with more than 250 employees, reported that they paid their male and female staff exactly the same, that is they had a zero average gender pay gap measured by both the mean and median. Experts on pay said that it was highly anomalous for companies of that size to have median and mean pay gaps that were identical because the two statistics measure different things. The mean gap measures the difference between the average male and female salary while the median gap is calculated using the midpoint salary for each gender.

How excited would you be about a medication that lowered your risk of cardiovascular death, heart attack or stroke by 1.5%? Excited enough to spend a few thousand dollars a year on the drug? I expect not. What if, instead, the drug reduced those same terrible outcomes by 20%? That's probably enough benefit to interest some in the drug. Well, those statistics come from the same clinical trial, evaluating the same drug. In fact, they present the exact same results, but they simply do it in different ways.

Short (10 minute) podcast "How to question dubious statistics in just a few short steps."....it's got some very British humour (sorry), but pretty good!

Some advice on evaluating claims made on data Bergstrom believes that calling bullshit on data, big or otherwise, doesn't require a statistics degree-only common sense and a few habits of mind. "You don't have to understand all the gears inside a black box in order to evaluate what you're being told," he said. For those who were unable to enroll in INFO 198/BIOL 106B this spring, here is some of his and West's advice:

McDonald's Monopoly competition is back this month offering a chance to win expensive prizes, all for the price of a Big Mac. Given you could become tens of thousands of dollars richer by simply going on a Macca's run, McDonald's Monopoly games have in the past been subject to cheating and a multimillion-dollar scandal. But for those who prefer to play fair, what are your chances of actually snaring a prize?

It's human nature to spot patterns in data. But we should be careful about finding causal links where none may existStories about people getting blood clots soon after taking the Oxford/AstraZeneca vaccine have become a source of anxiety among European leaders. After a report on a death and three hospitalisations in Norway, which found serious blood clotting in adults who had received the vaccine, Ireland has temporarily suspended the jab. Some anxiety about a new vaccine is understandable, and any suspected reactions should be investigated. But in the current circumstances we need to think slow as well as fast, and resist drawing causal links between events where none may exist.