Training Statistics Teachers at Iowa State University
Froelich, Amy G., Duckworth, William M., Stephenson, W. Robert, The American Statistician
The process of training graduate students to be statistics teachers is an informal one in the Department of Statistics at Iowa State University. We do not teach a course for graduate students in methods of teaching statistics. However, we do have a specific process for selecting and mentoring graduate student teaching assistants. First-year graduate students start as laboratory instructors/graders where they facilitate hands-on group activities for undergraduate students in introductory statistics classes. Those graduate students who do well as laboratory instructors are given the opportunity to teach a section of an introductory statistics course in their second year. The department provides mentoring and resources to help graduate student instructors meet the challenges of teaching statistics to undergraduate students.
2. INTRODUCTORY STATISTICS COURSES
Before discussing the role of graduate students in instruction in the Department of Statistics at Iowa State, it would be helpful to give some background to the reader about the structure of the introductory courses in the department. We offer several different introductory courses, each designed for a particular group of undergraduate majors. Statistics 101 is the general introductory course for all nonbusiness or nonengineering majors. This course meets three times (50 minutes each) a week in a large (approximately 100 students) "lecture" section. Lecture is in quotes because these meetings include small group activities, demonstrations, and opportunities for students to actively participate in the class. The class is split into two laboratory sections (approximately 50 students each) for a two-hour laboratory each week. The laboratory sessions consist of hands-on activities including data collection, data analysis, random sampling, designing experiments, and simulation activities on the central limit theorem, confidence intervals, and the importance of randomization as a basis for inference. The laboratory period is also used to review for, or give, exams and for students to work on a group data collection and analysis project.
Statistics 104 is the introductory course for agriculture and biology majors. This course meets twice a week for "lecture" (50 minute periods) and once for a two-hour laboratory. The laboratory is similar to that for Stat 101 but with examples and activities with an agricultural or biological context. The third introductory level statistics course is for business majors, Stat 226. This class meets three times a week in 50 minute "lecture" sections. There is no laboratory component to this course. (There are several different introductory courses for engineering majors, however, these courses are not often taught by graduate students.)
Each semester there are five lecture sections each of Stat 101, 104, and 226 with approximately 100, 50, and 80 students in each section, respectively. A faculty member, called the course coordinator, is in charge of each course. In a given semester, the course coordinator usually teaches one section of the introductory course and the remaining sections are taught by graduate students. Graduate students are thus responsible for teaching approximately 80% of the students enrolled in these courses.
Over the years, the introductory courses have evolved in response to the recommendations of the statistics reform movement to include more data, more concepts, and more use of the computer. In particular, we have tried to incorporate the suggestions of Moore (1997) in terms of reforming the content and pedagogy in the introductory courses. The primary goals of each introductory statistics course are to have students begin to understand statistical thinking and to be able to apply this understanding and to help answer substantive questions by collecting and analyzing appropriate data. To this end, each course now teaches a mix of methods (how to collect data, how to display data distributions, how to construct confidence intervals, etc. …