This interactive website from the BBC will match your student, using their height, gender, and weight, to their Rio Olympic body match. You enter your height, weight, age, and select your gender. It matches you with the athlete who is the most like you. It also provides good examples for distribution, and where you fall on the distribution, for Olympic athletes. I think it also gets students thinking about regression models. After you enter your data, the page returns information about where you fall on the distribution histogram for Olympic athletes by height, weight, and age for your gender. Then, the website returns your topic matches: How to use in class: 1) What other IVs could you collect to determine best sport match (DV)? Family income (I had access to soccer growing up, but not dressage horses)? Average temperature of hometown (My high school had a skiing club but not a beach volleyball club)? This gets your students thinking about multiple regression ...

As a research-parent-nerd joke before Christmas, six doctors swallowed Lego heads and recorded how long it took to pass the Lego heads. Why? As to inform parents about the lack of danger associated with your kid swallowing a tiny toy.  I encourage you to use it as a class example because it is short, it describes its research methodology very clearly, using a within-subject design, has a couple of means, standard deviations, and even a correlation. TL;DR: https://dontforgetthebubbles.com/dont-forget-the-lego/ In greater detail: Note the use of a within subject design. They also operationalized their DV via the SHAT (Stool Hardness and Transit) scale. *Yeah. So here is the Bristol Stool Chart  mentioned in the above excerpt. Please don't click on the link if your are eating or have a sensitive stomach. Research outcomes, including mean and standard deviations: An example of a non-significant correlation, with the SHAT score on the y-axi...

Maki Naro created a terrific comic strip detailing the replication, where it came from, where we are, and possible solutions.  You can use it in class to introduce the crisis and solutions. I particularly enjoy the overall tone: Hope is not lost. This is a time of change in statistics and methodology that will ultimately make science better. A few highlights: *History of science, including the very first research journal (and why the pressure to get published has lead to bad science) *Illustration of some statsy ways to bend the truth in science  *References big moments in the Replication Crisis  *Discusses the crisis AND solutions (PLOS, SIPS, COS)

They are bar graphs. And they are funny.

My friends. For you, I have compiled all of my funny statistic-Christmas images into one Google Slide presentation. You are welcome:  https://docs.google.com/presentation/d/12nmMw-69Ez71VmzaZ_QbUx4NOXiP17LHnvDjWUrx094/edit?usp=sharing

I don't find many examples of cluster analysis to share, but this example is REALLY engaging (using data to find serial killers), and is simple enough for a baby statistician BUT you can also make it a more advanced lesson as the data's owners freely share their data and code. Short Version: Journalist Thomas Hargrove (and his team) used cluster analysis to find clusters of similar killings within geographic areas. These might be a sign that a serial killer is active in that geographic region. It correctly identified a killer in Indiana. I found this interview from datainnovation.org which most succinctly describes the data analysis: https://www.datainnovation.org/2017/07/5-qs-for-thomas-hargrove-founder-of-the-murder-accountability-project/ Also statsy because the cluster analysis was validated using data from known serial killers. Hargrove's data and code can be accessed  here  and more information on his overall project to solve murders can be found...

UPDATE: 2020 Data: https://www.catalyst.org/knowledge/women-government When I explain chi-square at a conceptual, no-software, no-formula level, I use the example of gender distribution within the US Senate. There are 100 Senators, so the raw observed data count is the same as the observed data expressed via proportions. I think it makes it easier for junior statisticians to wrap their brains around chi-square.  I  usually start with an Goodness-of-Fit (or, as I like to call them, "One-sies chi-squares").For this example, I divide senators into two groups: men and women. And what do you get?  For the 115th Congress, there are 23 women and 77 men . There is your observed data, both as a raw count or as a proportion. What is your expected data? A 50/50 breakdown...which would also be 50 men and 50 women. Without doing the actual analysis, it is pretty safe to assume that, due to the great difference between expected and observed values, your chi-square Goodness o...