I maintain that relative versus absolute risk is a concept that we absolutely must teach in intro stats. I have given some examples of this before ( murder ! COVID !) but here is another one that hits home for my fellow Great Lakers. In particular, this one is for my friends in Detroit, Toledo, Cleveland, and Buffalo  Up here in Erie, a common point of discussion is how frozen Lake Erie is. Because once it freezes, the lake's moisture no longer feeds Dread Lake Effect Snow.   I like this example because you can easily perform the math in front of your class, demonstrating that 26.14% is 103.45% of 12.85%. At the same time, you have the visual to demonstrate that the vast majority of the lake was still unfrozen even with a 103.45% increase. 

I can't be the only one teaching online statistics this Spring. Last fall, I refreshed ALL of my discussion boards for my online version of Psychological Statistics. I haven't done so since 2020, and my students responded well to my new discussion topics, all of which are centered in statistical literacy and improving problem-solving with data. My first one is based on this old blog post about how residents of Houston used a Whataburger location map to figure out which parts of Houston were without electricity following Hurricane Meryl. Here is how I presented it to my students: You never know where valuable data visualizations will come from! For instance, following Hurricane Beryl, Texans used the Whataburger app to track power outages across Houston. Whataburger is a popular restaurant chain in the South. Its app has a feature where users can quickly see open or closed locations. Normally, this is used by hungry people to find the closest, open location. HOWEVER: There are S...

I know it is January 2025, but I want to write about something that happened during the Spring of 2024. I think it is a story about how it is never too late to do the right thing, making it great thing to think about here at the New Year. Data can't undo the past, but the way we manage it moving forward can provide the opportunity for some measure of equity. Back in May, professional baseball decided to include Negro League (NL), which existed from 2910 to 1948, baseball stats as part of Major League Baseball (MLB) stats. This is was done to allow for proper recognition of talented ML players. This changed some storied records for the league: https://www.mlb.com/news/stats-leaderboard-changes-negro-leagues-mlb This was a lot more than merging a couple of spreadsheets. As such, this story also serves as a lesson in data management and making desperate datasets the same. One that is a lot more moving than your typical story of data-cleaning. The following screenshots are from:  ...