The Anti-Frequentist Baby
It’s possible to educate Bayesian statisticians from infancy. Eric-Jan Wagenmakers, of JASP fame, shows us how with his new book, Bayesian Thinking for Toddlers. Continue reading The Anti-Frequentist Baby
Category: statistics
Not-Quite-in-Time Logistics
Writing in the Atlantic, Helen Lewis explains shortages and panic buying as a failure of efficiency; the weakness of just-in-time logistics. Sorry Helen, but you’re a logistics n00b; even this old retired Lieutenant Colonel knows about arcane loggie stuff like stock levels and re-order points (zero is NOT a good re-order point). When the military does it right, it’s called readiness (the First Gulf War was “fought off the shop floor” — we were over-prepared) When a civilian does it right, he’s called a prepper (Thanks, Covid-19, for making America a nation of preppers. It’s about time). Lewis blathers on … Continue reading Not-Quite-in-Time Logistics
(Clinical) Trial of the Century
The World Health Organization steps up to the coronavirus plate with what appears to be history’s most ambitious screening experiment. On Friday, the World Health Organization (WHO) announced a large global trial, called SOLIDARITY, to find out whether any can treat infections with the new coronavirus for the dangerous respiratory disease. It’s an unprecedented effort—an all-out, coordinated push to collect robust scientific data rapidly during a pandemic. The study, which could include many thousands of patients in dozens of countries, has been designed to be as simple as possible so that even hospitals overwhelmed by an onslaught of COVID-19 patients … Continue reading (Clinical) Trial of the Century
Make Every Vote Count
Every semester, I begin my introductory biostatistics class with a simple “show of hands” experiment based on the Bouba-Kiki effect. Prior to the experiment I “volunteer” two students at random to count hands, and when hands are raised, each of my volunteers counts silently and independently. Invariably, the counts DO NOT MATCH. We briefly discuss the phenomenon of measurement error, and select from a set of alternatives (recounts, averaging, etc) to resolve the problem. It’s a perfect “teaching moment” that occurs spontaneously as a surprise to my students on the first day of class. I console them with the fact … Continue reading Make Every Vote Count
Tsoo Tsoon to be a Tsunami
Andrew Gillen at the Texas Public Policy Foundation says “Two Tsunamis are About to Hit Higher Education,” when :..the Department of Education released post graduate earnings and debt data broken down by college program — which will have a revolutionary impact on higher education.” A bit of poking around on the web gets you to the TPPF webpage College Earnings and Debt, which ranks hundreds of degree programs by median student debt and after-graduation income. It’s a nice interactive database, where you can compare programs among multiple colleges, or for a single college. GIllen touts this information (he calls it the … Continue reading Tsoo Tsoon to be a Tsunami
Counting is DIFFICULT
Eleven million? or 22 million? A new Yale/MIT study estimates the illegal alien population in the US somewhere in the range of 16.5 to 29.1 million (for us statisticians, that’s 22.8 ± 6.3 million). That’s a margin of error larger than the entire population of Los Angeles (3.99 million). Worse yet, this estimate suggests that the Census Bureau’s annual American Community Survey report of 11 million is a seriously low-ball estimate. The Center for Immigration Studies is in the low-ball camp, but their argumentum ab auctoritate seems a bit shrill, and unwilling to admit to the possibility of systematic bias … Continue reading Counting is DIFFICULT
Pick a Name at Random
Found a cool new tool useful in simulating data sets: the Random Name Generator. What a great way to fake up some data! I’ve been using it in a course that includes survey sampling. Continue reading Pick a Name at Random
Caught in the Draft
In December of 1969, the Selective Service held a lottery to determine the order in which young men would be called up for the Draft. My number was a low 53, and that set the course for much of my adult life. Turns out, the odds were against me. A nice description of what happened. More details available on Wikipedia. More details that you ever wanted to know. Was the 1970 lottery truly random? Some clever statistician at the College of the Redwoods shows a simple analysis with R that suggests men born in December (like me) got shafted. Want … Continue reading Caught in the Draft
News Flash! Men and Women are Different…
..despite the fact that some folks wish is wasn’t so Neurosexism: the myth that men and women have different brains https://t.co/OOtJaICQ0t — Gina Rippon (@ginarippon1) February 27, 2019 In fact, the National Institute of Health requires that sex be included as a variable in all studies: Sex can influence health & disease in many ways, which is why NIH requires that researchers consider sex as a biological variable (SABV) in all stages of research: https://t.co/G2fy2PxLrJ. Visit @NIH_ORWH for #SABV research tips. #WomensHealthInFocus #ThisIsNIH — NIH (@NIH) January 8, 2019 My favorite line from the review: “The mammalian brain is clearly … Continue reading News Flash! Men and Women are Different…
Past performance is no indication of future…
I am such a slow pony. I’ve just web-surfed my way into discovering Rob Hyndman’s Time Series Data Library, which has hundreds of time-series datasets suitable for every teaching need. I was looking for one of my old faves, from that hoary old classic, Forecasting, Time Series, and Regression, and voila! there it is. Continue reading Past performance is no indication of future…
The Home for Wayward Statisticians 