A poem reflecting on Type I errors and the use of the null hypothesis in testing by Micah Wascher, a high school student at North Carolina School of Science and Mathematics. The poem won an honorable mention in the 2025 A-mu-sing Contest.
A poem reflecting on Type I errors and the use of the null hypothesis in testing by Micah Wascher, a high school student at North Carolina School of Science and Mathematics. The poem won an honorable mention in the 2025 A-mu-sing Contest.
Lyrics by John Bailer from Miami University about the value of successful statistical modeling that won an honorable mention in the 2025 A-mu-sing Contest. The song is a parody of the 1972 hit song by Johnny Nash "I Can see Clearly Now" and the vocals on the audio were performed by Kent Peterson from Oxford Presbyterian Church in Miami Ohio. In using the song for teaching, John suggests that each verse could be connected to a question or two. Here are examples:
Verse 1: How does an effective visualization reveal pattern? What types of patterns might emerge? How might such patterns be captured in a model?
Verse 2: What are ways variables are 'engineered' as part of an analysis?
Verse 3: How does the introduction of a confounder in a model potentially impact the estimated coefficients of other variables in a model?
Verse 4: AUC* is one way the predictive quality of a model is described. What are other features of a useful model?
* AUC stands for Area Under the Curve for the Receiver Operator Characteristic
Bridge: Residuals following a normal distribution are expected for some types of models. What other distributions might be expected? What else might be learned from the residuals?
Verse 5: Why are we more confident with interpolations than extrapolations?
"P-value's More than Alpha" is a music video by David Yew, an undergraduate student at Singapore Management University, that reviews introductory normal theory testing. The music is a fun parody of Billy Joel's 1989 hit song "We Didn't Start the Fire" and took second place in the 2025 A-mu-sing Contest. David also credits his statistics instructor, Rosie Ching, for providing feedback.
A satirical song about data science written by Dick De Veaux from Williams College that received an honorable mention in the 2025 A-mu-sing Contest. The song lampoons arguments over control of the field of data science and its defining characteristics.
A cartoon to use in explaining how hypothesis testing typically includes a null hypothesis that nothing is going on except random chance and p-values are calculated under that assumption. The cartoon was created by Joy Reeves from the Rachel Carson Council of Duke University and took first place in the cartoon/joke category of the 2025 A-mu-sing competition.
A cartoon that can be used in discussing how choosing an appropriate sample size must balance budget and logistics along with statistical power. The cartoon was used in the April 2023 CAUSE cartoon caption contest and the winning caption was written by retired AP Statistics teacher Jodene Kissler. The cartoon was drawn by British cartoonist John Landers (www.landers.co.uk) based on an idea by Dennis Pearl from Penn State University. An alternate caption for the cartoon might be “The Negative Correlation Moving Company had trouble holding on to their shorter employees,” that can be used to discuss the difference between positive and negative associations.
A cartoon that can be a vehicle to discuss the value of approximations in statistical inference and the need to check the fit of models. The cartoon was used in the October 2022 CAUSE cartoon caption contest and the winning caption was written by Eric Vance, from University of Colorado in Boulder. The cartoon was drawn by British cartoonist John Landers (www.landers.co.uk) based on an idea by Dennis Pearl from Penn State University.
A cartoon that can be used to discuss the multiple testing issue and the concept of p-hacking. The cartoon was used in the June 2021 CAUSE cartoon caption contest and the winning caption was written by Jim Alloway from EMSQ Associates. The cartoon was drawn by British cartoonist John Landers (www.landers.co.uk) based on an idea by Dennis Pearl from Penn State University.
A cartoon that can be used to discuss the importance of using a paired analysis to reduce the variability in the response for a heterogeneous population. The cartoon was used in the February 2021 CAUSE cartoon caption contest and the winning caption was written by Jeremy Case from Taylor University.. The cartoon was drawn by British cartoonist John Landers (www.landers.co.uk) based on an idea by Dennis Pearl from Penn State University.
A cartoon providing a nice way to introduce the value of data mining for finding patterns in data but not as a gold standard for inference. The cartoon was used in the July 2020 CAUSE cartoon caption contest and the winning caption was written by Charles Eugene Smith from North Carolina State University. The cartoon was drawn by British cartoonist John Landers (www.landers.co.uk) based on an idea by Dennis Pearl from Penn State University.