So, this checks off a couple of my favorite requisites for a good teaching example: You can personalize it, it is contemporary and applicable, it illustrates a few different sorts of statistics.  Climate Central wrote this article about first frost dates, and how those dates, and an increasing number of frost-free days, create longer growing seasons.  The overall article is about how frosty the US is becoming as the Earth warms. They provide data about the first frost in a number of US cities. It even lists my childhood hometown of Altoona, PA, so I think there is a pretty large selection of cities to choose from. Below, I've included the screen grab for my current home, and the home of Gannon University, Erie, PA. The first frost date is illustrated with a line chart, but the chart also includes the regression line. Data for frosty, chilly Erie, PA The article also presents a chart that shows how frost is related to the length of the growing season in t...

Students love personalized, interactive stuff.  This website from Chirs Wilson over at Time allows your American students to enter their name and they recieve their British statistical doppleganger name in return. Or vice versa. And by statistical doppleganger, I mean that the author sorted through name popularity databases in the UK and America. He then used a Least Squared Error model in order to find strong linear relationships for popularity over time between names. How to use in class: Linear relationship LSE Trends over time

This tongue-in-cheek  regression analysis found a way to predict the "Hipster Music Index" of a given artist by plotting # of Facebook shares of said artist's Pitchfork magazine review on they y-axis and Pitchfork magazine review score on the x-axis. If an artist falls above the linear regression line, they aren't "hipster". If they fall below the line, they are. For example, Kanye West is a Pitchfork darling but also widely shared on FB, and, thus demonstrating too much popular appeal to be a hipster darling (as opposed to Sun Kill Moon (?), who is beloved by both Pitchfork but not overly shared on FB). As instructors, we typically talk about the regression line as an equation for prediction, but Priconomics uses the line in a slightly different way in order to make predictions. Also, if you go to the source article, there are tables displaying the difference between the predicted Y-value (FB Likes) for a given artist versus the actual Y-value, which coul...