(Inspired by past experience, past observations, and the complaints of undergraduates from three—count ‘em, three—universities.)

1. On the first day of class, go over the syllabus, ask if there are any questions, and then launch right into the material. Omit any form of road map for the course or any reasoning as to why everyone from bio majors to engineering majors to political science majors are required to take the course.
2. Pick the driest, most boring textbook possible. Use it and it alone as supplemental material to your lecture notes.
3. Assume that all students, regardless of actual background or major, are familiar with such thing as summation notation, factorials, slopes, and calculus. That sociology major sitting in the middle row who’s struggling simply because they don’t know that 5! = 5*4*3*2*1? That’s their problem, not yours.
4. Make sure your class is all application, no theory. After all, why would people need to understand the reasoning behind the tests they’re using so long as they know how to do the tests?
5. If you cannot implement the above method, try making sure your class is all theory and no application. After all, if you teach the theory really well, odds are the students will be able to derive the practical applications by themselves, right?
6. Stick to the most boring examples you can think of, and make sure every example you use is coming from the same area of research. Do you have a background in business? Every example should be business-related and involve as many technical terms as you can throw in there. Same idea if your background is biology or psychology. That way, students can really see how statistics can be used in practically every field—as long as that field is yours.
7. There are plenty of cool, funny, and downright fascinating examples where statistics are used in unique and exciting ways. Make sure you keep these engaging examples out of the classroom.
8. If there is a topic that the majority of the class is struggling with, assume that it’s their own faults for not studying it well enough and press onward to new material. Breadth, not depth, right? Who cares if there’s a section people are struggling with as long as you cover every chapter in the textbook by the end of the semester.
9. Offer only one explanation of each topic. All students learn the same, and thus why waste time trying to explain a concept in two or more different ways? If there is confusion over your explanation as to why we use an ANOVA versus a bunch of paired t-tests for comparing 3+ means, it’s not your problem. Everyone should just be able to understand your explanation with no problem, so long as they’re applying themselves.
10. Finally, lecture in the most unenthusiastic voice possible. After all, you’re talking about numbers, right? Numbers are obviously inherently boring and this boringness should be conveyed through your lecturing style. If students are willing to learn, they should be able to get past your droning voice at 8:30 in the morning. Those who fall asleep simply are slackers.

All of the above info must already be very obvious, because so many statistics teachers seem to glean their teaching techniques from at least one or two of the above points.

After all, if we’re not making statistics the most painful subject to learn, we’re not doing our jobs right.