Skip to main content

Posts

Latest Post:

Statistics hiding in plain sight: How this four-minute long interview on polling data can be used in class

Statistical literacy is important. I love examples of folks using statistical concepts without being snobby statistical gatekeepers.  For instance, see jargon-free statistical literacy that helps people make important decisions about their children's health or informs the  immigration debate .   As stats teachers, I think we are remiss if we don't talk about political polling in our classes. At least a little bit. Here is a brief, nonpartisan interview about political polling you can use in class to see whether students can identify the concepts they've learned. Here is a short  NPR i nterview o n  how to think properly about political polling. If you use the interview in class, you could ask your students to identify the statistics and research methods mentioned.  I clocked margin of error,  sampling error,  and research methodology evaluation. If using an audio interview isn't your thing, check out t his summary piece from NPR, based on the...

Fujii's "How to design effective scientific figures"

Ryosuke Fujii's (2026)  How to Design Effective Scientific Figures  lives up to its title and is written at a level accessible to undergraduate and graduate students when conference time rolls around.  It was published in Nature, and it focuses on understanding which type of figure to use depending on a) your data and b) your audience.  It is only three pages long, and I found myself nodding and agreeing with the main points, summarized here: Additionally, the article contains this helpful flowchart detailing when to use which graphs. I appreciate that the flowchart isn't jargon-heavy and that it focuses on functionality.  

What do SEMs and criminal investigations have in common?

I really, really believe in using non-scientific examples and comparisons to teach statistics. In this post, I will show you how I see comparisons between criminal investigations and scientific investigations. Specifically, after a summer spent going down a true-crime rabbit hole, I think that criminal investigations and court cases are sort of like SEM. 1. Investigators and statisticians are sometimes forced to use tangible evidence to infer about intangible variables and concepts. 2. An evidence board is basically a path diagram, right? See: Don't worry, I made a whole PowerPoint to explain . And maybe you can use the PowerPoint the next time you teach SEM. If this idea was helpful, you'll probably enjoy my textbook, Psychological Statistics for Everyone . It's basically this blog, but organized into a semester.

Modern musician vocabularies: See how I extracted this data using GenAI, and how you can use it in class.

I intended for this to be a post about the singer vocabulary. It is still that, but it is also a post about using GenAI to grab data from an image. I mean, you can use Excel to do the same thing, but GenAI is a lot easier. Here we go. It starts with the Word Tips website, which helps you solve your crossword puzzles and Wordle. This website also has a blog dedicated to words. One such blog post explored which singers have the largest vocabularies, as measured by the number of  unique words in their lyrics. Their blog post compared music legends to newer talent. There are a ton of fun data visualizations on the website; go check it out. Since I teach college students, I decided to concentrate on the musicians my students listen to: In and of itself, this image serves as an example of bar graphs, good data visualization, and proper use of "buckets". However, I figured we could find a way to use the raw data in class. Create your own data visualization, create your own buckets....