For statistics classes in particular, the overwhelmingly common failure mode is: 1. Students learn how to apply some formulas to specific types of problems. I

For statistics classes in particular, the overwhelmingly common failure mode is: 1. Students learn how to apply some formulas to specific types of problems. If you ask them to find the population standard deviation of heights from a sample of 10 people, they can plug those numbers into their calculators and come up with the answer. 2. Except for maybe 10% of the class, they never really develop any deeper understanding. It's all surface level, "figure out which kind of problem this is then use the formula" stuff. 3. All this knowledge is forgotten very quickly once the grim specter of exams is gone. 4. Congratulations! "Education" has happened! The situation is less grim if e.g. you're teaching college students majoring in math or engineering, but would be worse if you tried pushing stats classes on a less selected group like the general high schooler population.