|
Abstract:
|
Beyond certain essential elements, both storytelling and applied statistics afford their practitioners creative liberty. As such, they are perhaps most easily taught in terms of what is unacceptable. In applied statistics courses, this is typically accomplished through project feedback. However, providing thoughtful, individualized feedback can be time-intensive for large classrooms, and students, having only seen the error in a specific context, may struggle to generalize the lessons. As an adjuvant to student learning and a time-saving device for instructors, we have compiled a list of the most common errors across dozens of student projects. To improve its usability, and to clarify how each error fits into the larger picture of statistical practice, we have organized the errors into a taxonomy with meaningful branch points related to the facets of a complete analysis. Our hope is that by studying the errors from the outset of a course, students will be primed to better understand the ways in which analyses fail. To facilitate generalization of insights, we provide multiple examples of each error appearing in a variety of contexts.
|
